U.S. patent application number 16/680213 was filed with the patent office on 2020-07-30 for virtual medical assistant methods and apparatus.
This patent application is currently assigned to Nuance Communications, Inc.. The applicant listed for this patent is Nuance Communications, Inc.. Invention is credited to Reid W. Coleman, Guido Gallopyn.
Application Number | 20200243186 16/680213 |
Document ID | 20200243186 / US20200243186 |
Family ID | 1000004780688 |
Filed Date | 2020-07-30 |
Patent Application | download [pdf] |
![](/patent/app/20200243186/US20200243186A1-20200730-D00000.png)
![](/patent/app/20200243186/US20200243186A1-20200730-D00001.png)
![](/patent/app/20200243186/US20200243186A1-20200730-D00002.png)
![](/patent/app/20200243186/US20200243186A1-20200730-D00003.png)
![](/patent/app/20200243186/US20200243186A1-20200730-D00004.png)
![](/patent/app/20200243186/US20200243186A1-20200730-D00005.png)
![](/patent/app/20200243186/US20200243186A1-20200730-D00006.png)
![](/patent/app/20200243186/US20200243186A1-20200730-D00007.png)
![](/patent/app/20200243186/US20200243186A1-20200730-D00008.png)
United States Patent
Application |
20200243186 |
Kind Code |
A1 |
Gallopyn; Guido ; et
al. |
July 30, 2020 |
VIRTUAL MEDICAL ASSISTANT METHODS AND APPARATUS
Abstract
In some aspects, a method of using a virtual medical assistant
to assist a medical professional, the virtual medical assistant
implemented, at least in part, by at least one processor of a host
device capable of connecting to at least one network is provided.
The method comprises receiving free-form instruction from the
medical professional, providing the free-form instruction for
processing to assist in identifying from the free-form instruction
at least one medical task to be performed, obtaining identification
of at least one impediment to performing the at least one medical
task, and inferring at least some information needed to overcome
the at least one impediment.
Inventors: |
Gallopyn; Guido; (Haverhill,
MA) ; Coleman; Reid W.; (Rehoboth, MA) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
Nuance Communications, Inc. |
Burlington |
MA |
US |
|
|
Assignee: |
Nuance Communications, Inc.
Burlington
MA
|
Family ID: |
1000004780688 |
Appl. No.: |
16/680213 |
Filed: |
November 11, 2019 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
13782438 |
Mar 1, 2013 |
10504622 |
|
|
16680213 |
|
|
|
|
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
G16H 40/20 20180101 |
International
Class: |
G16H 40/20 20180101
G16H040/20 |
Claims
1.-30. (canceled)
31. A host device for improving safety of performing one or more
medical tasks by providing access to a medical domain-specific
virtual clinical assistant, the host device being capable of
accessing at least one network, the host device comprising: at
least one microphone configured to receive free-form speech
instruction from a medical professional; at least one speaker
configured to output audio to the medical professional; and at
least one processor configured to: provide, to the medical
professional, at least one graphical user interface to the virtual
clinical assistant; using a dialog component, receive the free-form
speech instruction from the medical professional via the at least
one microphone; provide the free-form speech instruction for speech
processing, including automatic speech recognition, at least by
transmitting the free-form speech instruction over the at least one
network to an automatic speech recognition component to recognize
at least a portion of the free-form speech instruction; provide the
recognized free-form speech instruction to a natural language
understanding component to assist in automatically identifying from
the free-form speech instruction at least one medical task to be
performed; receive medical information relating to the free-form
instruction from a medical information resource via the at least
one network; and using the dialog component, establish a dialog
between the virtual clinical assistant and the medical professional
at least by: providing, via the at least one speaker, at least one
query spoken to the medical professional regarding the free-form
speech instruction prior to the at least one medical task being
performed; displaying the at least one query on the graphical user
interface; and receiving first information from the medical
professional, via the at least one microphone, responsive to the at
least one query to improve the safety of performing the at least
one medical task.
32. The host device of claim 31, wherein the free-form speech
instruction is processed by a natural language understanding
component to identify information expressed in the free-form speech
instruction indicating an intent to perform the at least one
medical task.
33. The host device of claim 32, wherein the at least one processor
is configured to obtain an identification of at least one
impediment to performing the at least one medical task, and wherein
the at least one query is provided to the medical professional
using the dialog component to facilitate overcoming the at least
one impediment.
34. The host device of claim 33, wherein the identified at least
one impediment includes one or more impediments selected from a
group consisting of: an identification of information missing from
the free-form speech instruction needed to perform the at least one
medical task, at least one ambiguity in connection with performing
the at least one medical task, an inability to identify the at
least one medical task with sufficient certainty, and one or more
conflicts in connection with performing the at least one medical
task.
35. The host device of claim 31, wherein the at least one medical
task identified from the free-form speech instruction includes a
request to order an orderable item.
36. The host device of claim 35, wherein the identified request to
order the orderable item includes an order selected from the group
consisting of: a medication order, a procedure order, a lab order,
a diagnostic test order, an appointment order, a referral order,
and a consultation order.
37. The host device of claim 36, wherein the identified at least
one impediment includes one or more conflicts including at least
one conflict between the orderable item and information identified
in the free-form speech instruction and/or patient information
obtained from an electronic medical record associated with the
patient, and wherein providing the at least one query comprises
providing an alert to the medical professional regarding the at
least one conflict using the dialog component.
38. The host device of claim 37, wherein the identified request to
order the orderable item includes a medication order for a
medication, and wherein the one or more identified conflicts
includes at least one conflict selected from the group consisting
of: a drug-drug interaction regarding the medication, a
drug-allergy interaction regarding the medication, a dose conflict
regarding the medication, a frequency of administration conflict
regarding the medication, and a route of administration conflict
regarding the medication.
39. The host device of claim 35, wherein the at least one processor
is configured to submit an order for the orderable item in response
to receiving the first information from the medical
professional.
40. The host device of claim 31, wherein the virtual clinical
assistant is hosted at least in part by at least one server
connected to the host device by the at least one network.
41. The host device of claim 40, wherein the at least one processor
is configured to communicate with the at least one server via the
at least one network in order to interface with one or more network
resources via the at least one network.
42. The host device of claim 31, wherein the at least one processor
is configured to: automatically identify at least one impediment to
performing the at least one medical task; and automatically infer
at least some information needed to overcome the at least one
impediment.
43. The host device of claim 42, wherein the at least some
information is inferred, at least in part, from Previously received
free-form speech instruction, information obtained responsive to
the Previously received free-form speech instruction, an electronic
medical record associated with a patient, and/or the medical
information resource accessible over the at least one network.
44. The host device of claim 36, wherein the at least one processor
is configured to: infer the at least some information at least by
inferring at least some information needed to complete the order of
the orderable item; and submit an electronic order for the
orderable item.
45. The host device of claim 36, wherein the identified request to
order the orderable item includes a request to order a medication,
and wherein inferring the at least some information includes
inferring at least one of a dose of the medication, a frequency of
administration of the medication, and a route of administration of
the medication.
46. The host device of claim 43, wherein a Previously received
free-form speech instruction comprises a request from the medical
professional to view patient information for a patient, and wherein
the at least one processor is configured to infer the at least some
information at least by inferring at least some information from
the patient information received responsive to the free-form speech
instruction and/or from an electronic medical record associated
with the patient identified in the free-form speech
instruction.
47. The host device of 42, wherein the at least some information is
inferred, at least in part, from Previously received free-form
speech instruction, information obtained responsive to the
previously received free-form speech instruction, an electronic
medical record associated with a patient, and/or the medical
information resource accessible over the at least one network.
48. A system for improving safety of performing one or more medical
tasks by providing access to a medical domain-specific virtual
clinical assistant, the system comprising: a host device configured
to access at least one network; and at least one server connected
to the at least one network, wherein the host device comprises: at
least one microphone configured to receive free-form speech
instruction from a medical professional; at least one speaker
configured to output audio to the medical professional; and at
least one processor configured to: provide, to the medical
professional, at least one graphical user interface to the virtual
clinical assistant, wherein the virtual clinical assistant is
hosted at least in part by the at least one server; using a dialog
component, receive the free-form speech instruction from the
medical professional via the at least one microphone; provide the
free-form speech instruction for speech processing at least in part
on the at least one server, including automatic speech recognition,
at least by transmitting the free-form speech instruction over the
at least one network to an automatic speech recognition component
to recognize at least a portion of the free-form speech
instruction; provide the recognized free-form speech instruction to
a natural language understanding component to assist in
automatically identifying from the free-form speech instruction at
least one medical task to be performed, wherein the at least one
server is configured to perform the speech processing including the
automatic speech recognition based on at least one statistical
model; receive medical information relating to the free-form
instruction from a medical information resource via the at least
one network; and using the dialog component, establish a dialog
between the virtual clinical assistant and the medical professional
at least by: providing, via the at least one speaker, at least one
query spoken to the medical professional regarding the free-form
speech instruction prior to the at least one medical task being
performed; displaying the at least one query on the graphical user
interface; and receiving first information from the medical
professional, via the at least one microphone, responsive to the at
least one query to improve the safety of performing the at least
one medical task.
49. A method comprising: processing a free-form input from a
medical professional, at least in part using natural language
understanding, to identify at least one medical task that the
free-form indicates is to be performed; determining, based at least
in part on the free-form input and/or the at least one medical task
determined in the processing, at least one query to be presented to
the medical professional regarding the at least one medical task;
outputting at least one query for presentation to the medical
professional; and receiving supplemental input from the medical
professional responsive, to the at least one query, to improve the
safety of performing the at least one medical task.
50. The method of claim 49, wherein determining the at least one
query based at least in part on the free-form input and/or the at
least one medical task comprises: determining an impediment to
performance of the at least one medical task; and determining a
query to solicit additional information from the medical
professional to aid in resolving the impediment.
Description
BACKGROUND
[0001] Many mobile communications devices, such as smart phones,
are equipped with a virtual assistant (also referred to as a
virtual agent) configured to receive speech input from a user and
employ various resources, either locally or accessible over a
network, to recognize the user's speech, attempt to understand the
user's intent and respond by performing one or more desired tasks
based on that understanding (e.g., perform an internet search, make
a telephone call, schedule an appointment, etc.). However,
conventional virtual assistants are configured to respond to voice
commands of the general user and therefore may be limited in their
capabilities, often focusing on general information retrieval,
personal scheduling or engaging communication functionality on the
mobile device (e.g., initiating a phone call, sending a text or
email message to an individual in the user's contact list,
etc.).
SUMMARY
[0002] Some embodiments include a method of using a virtual medical
assistant to assist a medical professional, the virtual medical
assistant implemented, at least in part, by at least one processor
of a host device capable of connecting to at least one network, the
method comprising receiving free-form instruction from the medical
professional, providing the free-form instruction for processing to
assist in identifying, from the free-form instruction, at least one
medical task to be performed, providing at least one first response
to the medical professional regarding the free-form instruction
prior to the at least one medical task being performed, and
receiving first information from the medical professional
responsive to the at least one first response.
[0003] Some embodiments include at least one computer readable
medium storing instructions that, when executed by at least one
processor, performs a method of using a virtual medical assistant
to assist a medical professional, the virtual medical assistant
implemented, at least in part, by at least one processor of a host
device capable of connecting to at least one network, the method
comprising receiving free-form instruction from the medical
professional, providing the free-form instruction for processing to
assist in identifying, from the free-form instruction, at least one
medical task to be performed, providing at least one first response
to the medical professional regarding the free-form instruction
prior to the at least one medical task being performed, and
receiving first information from the medical professional
responsive to the at least one first response.
[0004] Some embodiments include a host device capable of connecting
to at least one network, the host device comprising at least one
processor implementing, at least in part, a virtual clinical
assistant to assist a medical professional, the at least one
processor configured to, receive free-form instruction from the
medical professional, provide the free-form instruction for
processing to assist in identifying, from the free-form
instruction, at least one medical task to be performed, provide at
least one first response to the medical professional regarding the
free-form instruction prior to the at least one medical task being
performed, and receive first information from the medical
professional responsive to the at least one first response.
[0005] Some embodiments include a method of using a virtual medical
assistant to assist a medical professional, the virtual medical
assistant implemented, at least in part, by at least one processor
of a host device capable of connecting to at least one network, the
method comprising receiving free-form instruction from the medical
professional, providing the free-form instruction for processing to
assist in identifying from the free-form instruction at least one
medical task to be performed, obtaining identification of at least
one impediment to performing the at least one medical task; and
inferring at least some information needed to overcome the at least
one impediment.
[0006] Some embodiments include at least one computer readable
medium storing instructions that, when executed on at least one
processor, perform a method of using a virtual medical assistant to
assist a medical professional, the virtual medical assistant
implemented, at least in part, by at least one processor of a host
device capable of connecting to at least one network, the method
comprising receiving free-form instruction from the medical
professional, providing the free-form instruction for processing to
assist in identifying from the free-form instruction at least one
medical task to be performed, obtaining identification of at least
one impediment to performing the at least one medical task, and
inferring at least some information needed to overcome the at least
one impediment.
[0007] Some embodiments include a host device capable of connecting
to at least one network, the host device comprising at least one
processor implementing, at least in part, a virtual medical
assistant to assist a medical professional, the at least one
processor configured to receive free-form instruction from the
medical professional, provide the free-form instruction for
processing to assist in identifying from the free-form instruction
at least one medical task to be performed, obtain identification of
at least one impediment to performing the at least one medical
task, and infer at least some information needed to overcome the at
least one impediment.
BRIEF DESCRIPTION OF DRAWINGS
[0008] Various aspects and embodiments of the application will be
described with reference to the following figures.
[0009] FIG. 1 illustrates an exemplary system in which a virtual
medical assistant may be implemented, in accordance with some
embodiments;
[0010] FIG. 2 illustrates a method of interacting with a medical
professional to assist with one or more medical tasks specified in
free-form instruction from the medical professional, in accordance
with some embodiments;
[0011] FIG. 3 illustrates a method of inferring information in
connection with at least one medical task specified in free-form
instruction from the medical professional, in accordance with some
embodiments;
[0012] FIGS. 4A-4H illustrate examples of interaction between a
virtual medical assistant application installed on a mobile device
and a medical professional using the mobile device, in accordance
with some embodiments; and
[0013] FIG. 5 is an exemplary computer system on which techniques
described herein may be implemented, in accordance with some
embodiments.
DETAILED DESCRIPTION
[0014] As discussed above, conventional virtual assistants are
frequently adapted to be useful to a broad range of users and
therefore are generally limited to performing generic, and often
very few, tasks in response to user commands. Conventional virtual
assistants may therefore be unsuitable for performing specialized
tasks. The inventors have appreciated the benefit of a virtual
assistant adapted for the medical domain to assist a medical
professional in performing actions or tasks in the context of the
medical professional's medical practice and/or health care
capacity. Such a domain specific virtual assistant may be capable
of performing functionality within the medical domain that are
beyond the reach of conventional virtual assistants.
[0015] According to some embodiments, a virtual medical assistant
is implemented, at least in part, on a host device and is
configured to assist a medical professional in performing one or
more medical tasks including, but not limited to, assisting a
medical professional in ordering an orderable item (e.g., a
medication, a procedure, lab results, a diagnostic test, a
referral, a consultation, etc.), identifying conflicts, accessing
and displaying patient information, capturing dictation
corresponding to a patient encounter, updating an electronic
medical record (EMR) of a patient, scheduling a patient visit,
obtaining medical information, etc. Details of a virtual medical
assistant according to some embodiments are provided below.
[0016] The term "virtual medical assistant" refers herein to one or
more computer implemented components (e.g., software components,
hardware components, or a combination of both) configured to
receive input from a medical professional, initiate appropriate
processing of the input to identify one or more medical tasks
specified by the input and/or initiate one or more processes to
facilitate performance of the one or more identified medical tasks
specified by the input. A medical professional refers herein to a
person qualified in the practice of medicine, psychiatry, or
psychology, including, but not limited to, general physicians,
surgeons, psychiatrists, psychologists, therapists, or other
licensed health care professionals.
[0017] The inventors have recognized benefits of a virtual medical
assistant implemented, at least in part, on a host device operated
by a medical professional, wherein the virtual medical assistant is
configured to receive free-form instruction from the medical
professional and to facilitate processing of the free-form
instruction to identify and/or assist in performing one or more
medical tasks that the medical professional specified in the
free-form instruction. Free-form means that a medical professional
is generally unconstrained with respect to the structure and/or
content of the provided instruction. As such, a medical
professional may provide instruction using natural conversational
language that need not (but may) conform to any particular command
grammar and/or command lexicon, and have at least some instructions
understood by the virtual medical assistant. Accepting free-form
instruction allows a virtual medical assistant to be utilized
without requiring a medical professional to first learn a set of
structured commands to be understood by the virtual medical
assistant. The term "instruction" refers herein to any actionable
input including, but not limited to, one or more commands,
requests, queries, questions, inquiries, etc. Some examples of
free-form instruction are discussed in further detail below to
illustrate various concepts related to virtual medical assistant
methods and apparatus.
[0018] As discussed above, the inventors have recognized benefits
of a virtual medical assistant capable of receiving free-form
instruction and processing the free-form instruction to identify
one or more medical tasks specified in the free-form instruction.
Some medical tasks performed by a medical professional may be
relatively complicated in that they often require certain
information to be specified fully and in a particular manner in
order for the medical task to be performed. In addition, some
medical tasks relate to the health and well-being of a patient and
care must be taken to ensure the safety and suitability of certain
medical tasks to be performed. The inventors have appreciated that
a virtual medical assistant that can interact with a medical
professional via a dialog exchange may facilitate more efficient
performance of certain medical tasks. In addition, discourse
between the virtual medical assistant and the medical professional
may relieve some of the burden from the medical professional of
correctly and safely carrying out one or more medical tasks that a
medical professional would like performed.
[0019] According to some embodiments, a virtual medical assistant
is configured to receive free-form instruction from a medical
professional and initiate processing to identify one or more
medical tasks specified in the free-form instruction. The virtual
medical assistant may provide, via the host device or other
interface device, at least one response in connection with the
free-form instruction prior to the one or more medical tasks being
performed. The response provided by the virtual medical assistant
may include, for example, a request for additional information, a
request for clarification and/or disambiguation regarding the one
or more identified medical tasks, one or more alerts or warnings
with respect to a conflict in connection with the one or more
identified medical tasks, a confirmation regarding the one or more
identified medical tasks, an inquiry as to whether the medical
professional would like to proceed, or any other comment, question
or inquiry regarding the free-form instruction.
[0020] As discussed above, it may be inconvenient or time consuming
for a medical professional to specify every detail regarding a
medical task to be performed. In addition, some medical tasks
require certain information to be specified fully and in a specific
manner before the task can be performed, making it relatively
burdensome for a medical professional to accomplish certain medical
tasks. According to some embodiments, a virtual medical assistant
infers at least some information needed to perform one or more
medical tasks identified in free-form instruction from a medical
professional using information available to the virtual medical
assistant. For example, the virtual medical assistant may infer
information from one or more previously received free-form
instructions from the medical professional, information obtained in
performing one or more previous medical tasks, patient information
from an electronic medical record (EMR) associated with a patient
and/or information obtained from one or more medical information
resources accessible over a network.
[0021] Accordingly, a virtual medical assistant so configured may
be capable of assisting a medical professional in performing one or
more medical tasks specified in free-from instruction from the
medical professional, and may facilitate doing so even in the
presence of incomplete, ambiguous, inconsistent and/or conflicting
instruction from the medical professional by obtaining further
information from the medical professional and/or inferring
information based on data accessible to the virtual medical
assistant. The virtual medical assistant may also (or
alternatively) be configured to provide information to the medical
professional that may be useful to the medical professional in
completing one or more medical tasks, such as by identifying
conflicts and/or providing one or more alerts or warnings,
providing confirmation regarding a medical task to be performed,
etc., as discussed in further detail below.
[0022] Following below are more detailed descriptions of various
concepts related to, and embodiments of, methods and apparatus for
implementing a virtual medical assistant. It should be appreciated
that various aspects described herein may be implemented in any of
numerous ways. Examples of specific implementations are provided
herein for illustrative purposes only. In addition, the various
aspects described in the embodiments below may be used alone or in
any combination, and are not limited to the combinations explicitly
described herein.
[0023] FIG. 1 illustrates a system 100 within which a virtual
medical assistant may be implemented, according to some
embodiments. The virtual medical assistant comprises a virtual
medical assistant application 110 implemented, at least in part, by
one or more processors of a mobile device 101. Virtual medical
assistant application 110 may be installed as one or more software
components on mobile device 101, for example, by downloading the
virtual medical assistant application as an "app" via a network, or
otherwise installing the application on mobile device 101. Virtual
medical assistant application 110 may be configured to access and
utilize one or more network resources communicatively coupled to
the virtual medical assistant 110 via one or more networks 150
(e.g., via network communication capability available on the host
device).
[0024] Virtual medical assistant application 110 and one or more
network resources accessed by the virtual medical assistant
application 110 are collectively referred to herein as the virtual
medical assistant. Thus, actions described as being performed by
the virtual medical assistant are to be understood as being
performed by the virtual medical assistant application 110 and/or
any one or combination of network resources accessed, utilized or
delegated to by the virtual medical assistant application 110,
example resources of which are described in further detail below in
connection with the system illustrated in FIG. 1. Thus, according
to some embodiments, a virtual medical assistant may be implemented
as a distributed system having at least some functionality
implemented on a host device (e.g., mobile device 101) and at least
some functionality implemented via one or more network
resources.
[0025] In FIG. 1, the virtual medical assistant may be accessed by
a medical professional using mobile device 101 via virtual medical
assistant application 110 installed, at least partially, on the
mobile device as discussed above. Mobile device 101 in FIG. 1 is
illustrated as having a number of components characteristic of a
"smart phone," but it should be appreciated that mobile device 101
may be any suitable host device capable of receiving input from a
user (e.g., a medical professional), providing information (e.g.,
rendering information to a user visually, audibly, etc.), and
communicating with network resources via one or more networks, as
the aspects are not limited for use with any particular type of
host device.
[0026] In particular, in addition to mobile devices, suitable host
device types include personal computers or other generally
stationary or fixed computing devices accessible by the medical
professional. Another alternative to hosting a virtual medical
assistant includes one or more network accessible computers that
are accessed using telephony, wherein the network computer(s) hosts
the virtual medical assistant and the functionality is accessed via
a separate interface device. For example, a medical professional
may call into a system capable of accessing one or more host
devices using any suitable telephony device and conduct
interactions with the virtual medical assistant using speech, such
that the interface device (e.g., the telephony component used by
the medical professional) interacts with, but does not itself
necessarily host, the virtual medical assistant. Such a
telephony-based implementation may be accessed via the telephony
capabilities of a mobile device (e.g., using cellular technology),
or may be accessed using any other suitable telephony
component).
[0027] Mobile device 101 may include one or more transducers 103
for converting acoustic energy to electrical energy and vice versa.
For example, transducers 103 may include one or more speakers
and/or one or more microphones arranged on the mobile device to
allow input/output (I/O) of acoustic information. Different mobile
devices may include differing numbers and arrangements of
transducers, any of which may be suitable to facilitate acoustic
I/O. Mobile device 101 may include further I/O functionality
including display 105 for visually presenting information to the
user. Display 105 may also be configured as an input device, for
example, by responding to touch, pen input, etc. Mobile device 101
may include one or more other input components that allow a user to
interface with the mobile device. For example, button 102 may be
used as a selection button or may include any other functionality
on the mobile device. Mobile device 101 may include a keypad (not
shown), arrow buttons or any other input component capable of
receiving input from a user, such as a medical professional using
the mobile device to interact with virtual medical assistant
application 110.
[0028] Mobile device 101 may also include one or more wireless
communication components 106. For example, mobile communication
devices such as cellular telephones will typically have a wireless
transceiver capable of communicating with one or more cellular
networks. Alternatively, or in addition to, wireless communication
component 106 may include a wireless transceiver capable of
communicating with one or more other networks or external devices.
For example, wireless communication component 106 may include a
component configured to communication via the IEEE 802.11 standard
(Wi-Fi) to connect to network access points coupled to one or more
networks (e.g., local area networks (LANs), wide area networks
(WANs) such as the internet, etc.), and/or wireless communication
component 106 may include a Bluetooth.RTM. transceiver to connect
to a Bluetooth.RTM. compatible device, etc.
[0029] To simplify the illustration of exemplary mobile device 101,
wireless communication component 106 is used to represent
schematically any wireless communication component implemented on
mobile device 101 and may include one or multiple components that
may be either separate or integrated components, may be independent
system components, or may be different system components in
communication with one or more other wireless components. That is,
wireless communication component 106 denotes one or any combination
of components that may be implemented on a mobile device that
allows the mobile device 101 to communicate wirelessly with one or
more networks, systems and/or other devices. It should be
appreciated that mobile device 101 is merely an example of a mobile
device suitable for hosting a virtual medical assistant application
and/or for use with any one or combination of virtual medical
assistant techniques described herein.
[0030] Techniques described herein are not limited for use with any
particular device having any particular number or arrangement of
components, and any network aware device having I/O functionality
may be suitable to operate as the host device for a virtual medical
assistant application to access functionality of a virtual medical
assistant. As discussed above, for example, the host device may be
a personal computer or other standalone computer having network
access (e.g., either wired, wireless, or combination, access to one
or more networks), may be a network computer capable of being
accessed by telephony components, or any other suitable computing
device.
[0031] According to some embodiments, virtual medical assistant
application 110 is configured to present one or more interfaces to
receive input (e.g., free-form instruction from a medical
professional) via the I/O functionality of the host device (e.g.,
mobile device 101). For example, virtual medical assistant
application 110 may be configured to receive speech from a medical
professional via one or more microphones coupled to mobile device
101, and initiate processes to recognize and understand the content
of a medical professional's speech, as discussed in further detail
below. Alternatively, or in addition to, virtual medical assistant
application 110 may receive input from the medical professional in
other ways, such as via any of the input mechanisms provided on
mobile device 101 (e.g., touch sensitive display, keypad, mouse,
one or more buttons or any other interface control provided on the
host device). According to some embodiments, virtual medical
assistant application 110 is configured to provide a graphical user
interface (GUI) on display 105 of mobile device 101 that presents
to the medical professional the mechanisms by which the medical
professional may provide input to the virtual medical
assistant.
[0032] Virtual medical assistant application 110 may also be
configured to present one or more interfaces to provide information
to the medical professional. For example, virtual medical assistant
application 110 may display information to the user via the display
on the host device (e.g., via a GUI displayed on a display 105 on
mobile device 101). Virtual medical assistant application 110 may
also provide information audibly to the medical professional, for
example, using speech synthesis techniques and rendering
synthesized speech via one or more speakers coupled to mobile
device 101, or by rendering prerecorded speech or one or more
sounds configured to convey information to the medical
professional. According to some embodiments, virtual medical
assistant application 110 is capable of rendering information to
the medical professional both visually and audibly. Virtual medical
assistant application 110 may include other mechanisms for
providing information to the medical professional, as the aspects
are not limited for use with any particular type or technique for
providing and/or rendering information to the medical
professional.
[0033] According to some embodiments, a virtual medical assistant
may employ a virtual medical assistant application server 110'
coupled to network(s) 150 to communicate with virtual medical
assistant application 110 installed on mobile device 101 to assist
with interfacing with one or more network resources. By
implementing at least some functionality of the virtual medical
assistant on virtual medical assistant server 110', the computer
footprint of the virtual medical assistant application 110
installed on the host device may be reduced (e.g., on a mobile
device where resources may be limited to some extent). However, in
some embodiments, virtual medical assistant application server 110'
may not be implemented and functionality to interface with the one
or more network resources may be implemented by virtual medical
assistant application 110 resident on the host device, as the
aspects are not limited in this respect.
[0034] According to some embodiments, the virtual medical assistant
may not include a virtual medical assistant application installed
on the interface device used by the medical professional to access
the functionality of the virtual medical assistant. For example,
according to some embodiments, a medical professional may access
the virtual medical assistant via a telephony device that does not
host, or may be unsuitable for hosting, virtual medical assistant
application software, and may interact with the virtual medical
assistant via the telephony device. In such an implementation, the
virtual medical assistant application may be implemented by one or
more network computing devices accessible, either directly or
indirectly, by the telephony device (e.g., virtual medical
assistant application server '110 may implement all, or some
portion, of the virtual medical assistant application
functionality), and virtual medical assistant functionality may be
delivered by network components to the interface device used by the
medical professional.
[0035] According to some embodiments, virtual medical assistant
application 110 is configured to access, utilize and/or delegate to
one or more network resources coupled to network(s) 150, thus
forming a virtual medical assistant (denoted generally as virtual
medical assistant 100 in the example in FIG. 1) configured to
assist in performing medical tasks requested by or on behalf of a
medical professional. Network(s) 150 may be any one or combination
of networks interconnecting the various network resources
including, but not limited to, any one or combination of LANs,
WANs, the internet, private networks, personal networks, etc. The
network resources depicted in FIG. 1 are merely exemplary, and a
virtual medical assistant may comprise any one or combination of
network resources illustrated in FIG. 1, or may utilize other
network resources not illustrated, as techniques described herein
are not limited for use with any particular number or configuration
of network resources.
[0036] As depicted in FIG. 1, virtual medical assistant 100
includes a medical understanding resource 120 comprising automatic
speech recognition (ASR) component 130, natural language
understanding (NLU) and/or clinical language understanding (CLU)
component 140, and dialog component 160. Each of these components
may be implemented in software, hardware, or a combination of
software and hardware. Components implemented in software may
comprise sets of processor-executable instructions that may be
executed by one or more processors of one or more network
computers, such as a network server or multiple network servers.
Each of ASR component 130, NLU/CLU component 140 and dialog
component 160 may be implemented as a separate component, or any
combination of these components may be integrated into a single
component or a set of distributed components implemented on one or
multiple network computers (e.g., network servers). In addition,
any one of ASR component 130, NLU/CLU component 140 and dialog
component 160 may be implemented as a set of multiple software
and/or hardware components. As such, processing described as being
performed by medical understanding resource 120 refers to
processing performed by any one or combination of the constituent
components, operating alone or in concert with other constituent
components.
[0037] The system illustrated in FIG. 1 also comprises a number of
medical resources including electronic health record (EHR) system
170, medical decision support (CDS) service 180, order fulfillment
system 190 and medical information retrieval services 155, each of
which are discussed in further detail below. It should be
appreciated that the network resources illustrated in FIG. 1 are
merely exemplary and a virtual medical assistant may utilize one or
any combination of resources illustrated, or may utilize other
resources not illustrated or described. It should be further
appreciated that the various components illustrated in FIG. 1 may
be coupled in any suitable manner, and may be components that are
located on the same physical computing system(s) or separate
physical computing systems that can be coupled in any suitable way,
including using any type of network, such as a local network, a
wide area network, the internet, etc.
[0038] As discussed above, virtual medical assistant application
110 may receive input (e.g., free-form instruction from a medical
professional) in the form of speech, and ASR component 130 may be
utilized to identify the content of the speech (e.g., recognize the
constituent words in the speech input). In particular, a medical
professional may speak free-form instruction to the host device
(e.g., mobile device 110) and/or an interface component capable of
accessing, directly or indirectly, the host device. Virtual medical
assistant application 110 may receive the free-form instruction
from the medical professional and provide the free-form instruction
to ASR component 130 to be recognized. The free-form instruction
may be processed in any suitable manner prior to providing the
free-form instruction to ASR component 130. For example, the
free-form instruction may be pre-processed to remove information,
format the free-form instruction or modify the free-from
instruction in preparation for ASR (e.g., the free-form instruction
may be formatted to conform with a desired audio format and/or
prepared for streaming as an audio stream or prepared as an
appropriate audio file) so that the free-form instruction can be
provided as an audio input to ASR component 130 (e.g., transmitted
over a network).
[0039] ASR component 130 may be configured to process audio input
(e.g., audio input representing free-form instruction) to form a
textual representation of the audio input (e.g., a textual
representation of the constituent words in the free-form
instruction that can be further processed to understand the meaning
of the constituent words). Such processing to produce a textual
representation may be performed in any suitable way. In some
embodiments, ASR component 130 may convert speech to a
representation other than a textual representation, or the speech
may not be recognized as words, but instead a sequence or
collection of abstract concepts. It should be appreciated that the
words and/or concepts represented in the medical professional's
free-form instruction may be represented and/or stored as data in
any suitable form, as aspects are not limited in this respect. ASR
component 130 may transmit the recognized free-form instruction
over the network to be received by the virtual medical assistant
application 110 (e.g., for display to the medical professional)
and/or ASR component 130 may provide the recognized free-form
instruction to one or more other network resources for further
processing (e.g., the recognized free-form instruction may be
provided to a NLU/CLU component 140 to assist in understanding the
semantic content of the free-form instruction to identify one or
more medical tasks specified therein), as discussed in further
detail below.
[0040] According to some embodiments, ASR component 130 may be
tailored, at least in part, to the medical domain. For example, ASR
component may make use of a lexicon of medical terms (which may be
part of, or in addition to, another more general speech recognition
lexicon) while determining the sequence of words that were spoken
in the free-form instruction from the medical professional.
However, aspects of the invention are not limited to the use of a
lexicon, or any particular type of lexicon, for ASR. When used, the
medical lexicon in some embodiments may be linked to a knowledge
representation model (e.g., a medical language understanding
ontology, one or more statistical models and/or one or more
rule-based models) utilized by NLU/CLU component 140 and/or dialog
component 160, such that ASR component 130 produces a textual
representation of the free-form instruction containing at least
some terms in a form understandable to NLU/CLU component 140 and/or
dialog component 160.
[0041] In some embodiments, ASR component 130 may use a more
general speech recognition lexicon shared between ASR component and
one or more other network resources. However, in other embodiments,
ASR component 130 may not have any lexicon developed to be in
common with the other exemplary network resources illustrated in
FIG. 1 or otherwise described. In some embodiments, a lexicon used
by ASR component 130 may be linked to a different type of medical
knowledge representation model, such as one not designed or used
for language understanding. It should be appreciated that any
lexicon used by ASR component 130 (or any of the other network
resources) may be implemented and/or represented as data in any
suitable way. ASR component 130 may utilize one or any combination
of suitable acoustic models, lexicons, grammars, language models,
etc. to perform recognition, as aspects are not limited for use
with any particular ASR implementation.
[0042] As discussed above, the medical understanding resource 120
may include NLU/CLU component 140 configured to process free-form
instruction to identify one or more medical tasks specified in the
free-form instruction and may also identify one or more parameters
associated with the one or more medical tasks. According to some
embodiments, a virtual medical assistant is configured to process
free-form instruction from the medical professional. To this end,
NLU/CLU component 140 may be designed to be relatively robust with
respect to its ability to identify tasks in free-form instruction
provided in as many of the myriad ways in which a medical
professional might choose to express the instruction to the extent
desired or to the extent reasonable (or possible). However, there
is no limitation on what expressions an NLU/CLU component 140 may
be capable of understanding, as implementations according to some
embodiments may be more limited than others in this respect.
[0043] NLU/CLU component 140 may use any suitable language
understanding techniques to ascertain the meaning of the free-form
instruction so as to assist in identifying one or more tasks
specified therein. For example, NLU/CLU component may be configured
to identify and extract grammatical and/or syntactical components
of the free-form speech, such as parts of speech, or words or
phrases belonging to known semantic categories, to facilitate an
understanding of the free-form instruction. For example, NLU/CLU
component 140 may identify action words (e.g., action words that a
medical profession may be inclined to use in specifying certain
medical task), subject words, topic words, medical terms, clinical
facts and/or any other type or category of words the NLU/CLU
component 140 may deem relevant to ascertaining the semantic form
or content of the free-form instruction to facilitate identifying
medical tasks specified therein.
[0044] To assist in identifying medical tasks in free-form
instruction, NLU/CLU component 140 may have specialized
functionality tailored to the medical domain. For example, NLU/CLU
component 140 may include one or more knowledge representation
models that captures semantic knowledge regarding language in
general and/or medical language in the specific. For example, a
knowledge representation model may be capable of associating
medical terms with corresponding categories, classifications or
types. Such categories may include, but are not limited to,
medications, allergies, procedures, diagnostic tests, laboratory
tests, conditions, disease or disorders, symptoms, vital signs,
anatomy, body sites, medical devices, observable findings,
measurable findings, etc. It should be appreciated that the above
list is merely exemplary, and some embodiments may not use one or
more of the above-listed categories, and some embodiments may use
other suitable categories.
[0045] A knowledge representation model may make use of or
incorporate a formal ontology linked to a lexicon of medical terms.
The formal ontology may be implemented as a relational database, or
in any other suitable form, and may represent semantic concepts
relevant to the medical domain, as well as linguistic concepts
related to ways the semantic concepts may be expressed in natural
language. Medical terms in the lexicon may be linked to at least
one concept in the formal ontology.
[0046] A knowledge representation model may use, employ or
implement one or more statistical models to extract semantic
entities from natural language input. In general, a statistical
model can be described as a functional component designed and/or
trained to analyze new inputs based on probabilistic patterns
observed in prior training inputs. In this sense, statistical
models differ from "rule-based" models, which typically apply
hard-coded deterministic rules to map from inputs having particular
characteristics to particular outputs. By contrast, a statistical
model may operate to determine a particular output for an input
with particular characteristics by considering how often (e.g.,
with what probability) training inputs with those same
characteristics (or similar characteristics) were associated with
that particular output in the statistical model's training data.
For example, in the context of a virtual medical assistant, the
input on which the model operates may correspond to free-form
instruction (or a representation derived therefrom) and the output
may include an identification of a medical task specified by the
free-form instruction and/or one or more associated parameters.
[0047] To supply the probabilistic data that allows a statistical
model to extrapolate from the tendency of particular input
characteristics to be associated with particular outputs in past
examples, statistical models are typically trained (or "built") on
training corpuses with relatively large numbers of example inputs.
Typically, the example inputs are labeled with the known outputs
with which they should be associated (e.g., a corpus of free-form
instruction input may be labeled with the correct medical task the
instruction specifies), often by a human labeler (e.g., with expert
knowledge of the domain), or sometimes automatically.
Characteristics of interest (known as "features") are identified
("extracted") from the inputs, and the statistical model learns the
probabilities with which different features are associated with
different outputs, based on how often training inputs with those
features are associated with those outputs. When the same features
are extracted from a new input (e.g., an input that has not been
labeled with a known output, for example, by a human), the
statistical model can then use the learned probabilities for the
extracted features (as learned from the training data) to determine
which output is most likely correct for the new input. In the
context of a virtual medical assistant, the features may correspond
to the words, concepts, associations, relationships, etc.
identified in the free-form instruction, and the statistical model
may determine which one or more tasks is most likely correct for
free-form speech received from the medical professional based on
these features and their relationships.
[0048] A knowledge representation model utilized by NLU/CLU
component 140 may also include (or alternatively include) one or
more rule-based models that provide a set of rules as to how to map
words or phrases in free-form instruction to corresponding medical
tasks and/or that map words or phrases in free-form instruction to
parameters of an identified medical task. For example, NLU/CLU
component 140 may include a rule-based natural language processing
system to extract medical facts, link facts to concepts or
otherwise assist in identifying one or more medical tasks specified
in free-form instruction received from a medical professional. In a
rule-based system, a linguist and/or other individual may create a
plurality of rules that can specify what words or combinations or
words evidence that free-form speech specifies a particular medical
task (e.g., an instruction to order an orderable item). Rule-based
techniques can be employed by NLU/CLU component 140 to identify one
or more medical tasks specified in free-form instruction from the
medical professional (e.g., a request for patient information,
instruction to place an order for an orderable item, instruction to
take dictation, command to update a patient record, inquiry
regarding a medical entity, etc.) and/or any parameters associated
with identified medical tasks.
[0049] It should be appreciated that an NLU/CLU component may
utilize one or any combination of techniques described above to
identify medical tasks in free-form instruction and/or identify any
parameters associated with identified medical tasks, as aspects are
not limited for use with any particular technique or combination of
techniques. In accordance with one embodiment described herein, a
hybrid approach is used in which NLU/CLU component 140 employs both
a statistical model and a rules-based component to identify one or
more medical tasks specified in free-form instruction. Such a
hybrid approach can be implemented in any of numerous ways, as
aspects that employ a hybrid approach are not limited to any
particular implementation technique. In accordance with some
embodiments, the statistic model and rules-based component of the
NLU/CLU component 140 may operate in parallel on the same free-form
instruction, and in some embodiments they may operate serially. In
either parallel or serial operation, the statistic model and the
rules-based component may operate independently or may exchange
information to facilitate identifying medical task(s) specified in
free-form dictation, and/or parameters associated with a medical
task.
[0050] According to some embodiments, a rules-based component may
perform a first pass on free-form instruction to identify semantic
entities by, for example, extracting medical facts, linking words
or phrases into concepts, and/or identifying other characteristics
of known medical tasks. Thereafter, the statistical model may be
used to process the semantic entities to determine what one or more
medical tasks the free-form instruction most likely relates to.
That is, the outputs of the rules-based component can be used as
features for the statistical model. For example, when a rule fires
or otherwise indicates that the rule has been met, this may be a
feature evaluated by the statistical model. Some rules may be quite
specific, so that a firing of the rule indicates with high
probability that the determination expressed thereby is accurate
(e.g., detection of the word "order" in combination with
identifying the name of a medication in the free-form instruction
may indicate with high probability that the medical professional
would like the virtual medical assistant to order the
medication).
[0051] According to some embodiments, a statistical model may
operate to identify the one or more medical tasks most likely
specified by the free-form instruction and the rule-based component
may be used to identify one or more parameters from the free-form
instruction associated with the identified medical task. A
statistical model and a rule-based component may be used in other
suitable ways, as the aspects are not limited in this respect. In
accordance with some embodiments, a determination can be made about
whether to use a statistical model or a rule-based component
depending upon the content of the free-form instruction. For
example, some instructions that are very structured may lend
themselves to rules, whereas types of instructions with significant
variability in how they are expressed in natural language may be
better suited to the use of a statistical model.
[0052] In accordance with some embodiments, an ontology of the type
described above can be used in conjunction with a statistical model
in any of numerous ways. For example, the use of an ontology can
reduce the amount of annotated data that can be used in training
the statistical model, as the ontology can demonstrate a number of
alternate ways of expressing the same concept. Thus, it may not be
necessary to include annotated data representing all of the
alternate ways of expressing instruction to perform certain medical
tasks, and the ontology can be combined with the statistical model
to expand its ability to recognize alternative expressions not
included in the annotated training data. Of course, this is merely
one use of an ontology, as the ontology can be used in any suitable
manner.
[0053] Again, the descriptions provided above for combining the use
of a rule-based component with a statistical model and optionally
with an ontology are merely illustrative, as the techniques
described herein are not limited to any particular implementation
techniques. The techniques described herein can be implemented
using solely a statistical model, solely a rule-based component, or
any combination, as they are not limited to any particular
implementation technique. That is, processing by NLU/CLU component
140 may be as simplistic or as sophisticated as desired to suitably
identify one or more medical tasks when they are present in
free-form instruction provided by a medical professional.
[0054] As mentioned above, medical understanding resource 120
illustrated in FIG. 1 also may include dialog component 160. Dialog
component 160 may include conversational functionality that
facilitates discourse between a medical professional and the
virtual medical assistant. As discussed above, the virtual medical
assistant may be configured to provide one or more responses to a
medical professional in connection with free-form instruction
received from the medical professional. Dialog component 160 may
contribute, in full or in part, to the formulation of such
responses based on the content of the free-form instruction, the
context of the situation, and/or the intent of the one or more
responses to be provided to the medical professional.
[0055] Dialog component 160 may be coupled to NLU/CLU component 140
so as to obtain information regarding the content of the free-form
instruction (e.g., to receive information regarding one or more
identified medical tasks specified in the free-form instruction) to
ascertain whether a response to the medical professional is needed
and, if so, how the response should be formulated and what
information the response should contain. Dialog component 160 may
be capable of carrying out different dialogs as appropriate to
obtain information needed to complete a plurality of medical tasks.
For example, a medical task may have an associated structured
representation comprising a plurality of fields corresponding to
information regarding the respective medical task that is
maintained by dialog component 160 or by a component accessible by
dialog component 160. Some fields may correspond to parameters that
are required to perform the associated medical task and some fields
may correspond to optional parameters.
[0056] Dialog component 160 may operate in conjunction with NLU/CLU
component 140 to identify one or more medical tasks specified in a
medical professional's free-form instruction, and to populate the
appropriate fields in the representation associated with the
identified medical task(s) either with information extracted by
NLU/CLU component 140 from the free-form instruction, information
inferred therefrom, inferred from the context or state of the
discourse with the medical professional, or inferred from data
available from one or more network resources. Dialog component 160
may determine that one or more required fields have not been
populated as a result of processing received free-form instruction
(e.g., certain information could not be obtained or inferred from
the free-form instruction, or otherwise inferred) and, in response,
formulate an appropriate response to the medical professional to
obtain the missing information. The response may then be provided
to the medical professional via the virtual medical assistant
application 110 on the host device (e.g., the medical
professional's mobile device). Similarly, dialog component 160 may
ascertain that certain optional information is missing and deem it
appropriate to formulate a response designed to solicit this
information from the medical professional. Dialog component 160 may
also be configured to formulate a response when the information
needed to perform a medical task is complete, for example, to
provide a confirmatory response to the medical professional to
ensure that the task is the one intended and the parameters
ascertained by the virtual medical assistant are correct.
[0057] Dialog component 160 may be configured to work in concert
with CLU/NLU component 140 to formulate responses designed to
solicit information from the medical professional to resolve any
ambiguities and/or uncertainty that resulted from processing
free-form instruction received from the medical professional. For
example, processing performed by CLU/NLU component 140 may have
resulted in uncertainty with respect to the medical task the
medical professional would like performed or may have resulted in
multiple possibilities in this respect. Processing by CLU/NLU
component 140 may have resulted in uncertainty regarding one or
more parameters associated with the medical task or needed to
complete the medical task. Such ambiguity and/or uncertainty may
have resulted from an incomplete understanding of the free-form
instruction, incomplete or relatively low confidence recognition of
the free-form instruction (e.g., when the free-form instruction is
received as speech), incomplete or ambiguous instruction from the
medical professional, or a combination thereof. CLU/NLU component
140 may convey any ambiguity and/or uncertainty to dialog component
160 to formulate a response designed to resolve any
ambiguity/uncertainty or obtain clarification regarding questions
or concerns in connection with the free-form instruction.
[0058] As discussed above (and as discussed in further detail
below), the virtual medical assistant may be configured to process
free-form instruction and identify one or more conflicts with
respect to the compatibility and/or suitability of performing one
or more medical tasks identified from free-form instruction
received from the medical professional. For example, problematic
interactions between one or any combination of medications,
allergies, procedures, conditions, etc., may be identified in the
course of processing the free-form instruction. As discussed in
further detail below, such conflicts may be identified in part by
utilizing one or more medical resources such as EHR system 170, CDS
resource 180, medical information retrieval resource 155 or other
available resources. Dialog component 160 may be configured to
receive information regarding such conflicts and formulate one or
more responses to alert the medical professional as to the
existence of the conflict and/or to inquire how the medical
professional would like to proceed in view of the identified
conflict(s).
[0059] According to some embodiments, dialog component 160 may
assess whether a compound instruction is appropriate under given
circumstances. For example, based on the assessment that the
medical professional is intending to order a medication, dialog
component 160 may determine that it is important to both recite the
identified parameters that have been extracted from the free-form
instruction (e.g., by NLU/CLU component 140) or inferred therefrom,
as well as inquire after one or more missing parameters.
[0060] The above described functionality with respect to dialog
component 160 is merely exemplary and a given dialog component may
implement some portion or all of the functionality described above,
or may implement additional functionality not described herein or
described in connection with another component or resource. In
addition, some of the functionality described in connection with
dialog component 160 may be performed, in whole or in part, by
other components or resources (including the virtual medical
assistant application resident on the host device and/or a virtual
medical assistant application server, if implemented), as
functionality described above, when and if implemented, can be
performed by any suitable component or combination of
components.
[0061] As discussed above, the system illustrated in FIG. 1 also
includes a number of medical resources, including EHR system 170.
EHR system 170 may include one or more components that store the
electronic medical records for patients of one or more health care
institutions. Many healthcare institutions are transitioning or
have transitioned from paper documentation to electronic medical
record (EMR) systems, in which patients' medical information is
stored in a data repository in electronic form. The information
that was previously memorialized in paper form as the patient's
"chart" may be archived as an electronic health record (EHR) for
the patient, along with any other relevant patient data. For
example, an EHR for a patient may include any one or combination of
the patient's demographic information, intake forms, history,
immunization records, laboratory results, medical professional's
notes memorializing patient encounters/visits, surgical, pathology
and/or radiology reports, diagnostic reports, imaging results or
other diagnostic test results, consult notes, discharge summaries,
etc.
[0062] Thus, an EHR is an electronic medical record that generally
is maintained by a specific healthcare institution and contains
data documenting the care that a specific patient has received from
that institution (or other institutions) over time. EHR system 170
may include electronic medical records for patients from a given
institution or from multiple institutions. EHR system 170 may
include one or any combination of separate EHR systems,
corresponding to respective single or multiple institutions, or may
include one or any combination of EHR systems that are integrated
or otherwise communicatively coupled. EHR system 170 may also
include other functionality in addition to archiving, managing and
updating electronic medical records and/or may store other medical
information of a more general nature (e.g., not specific to a
particular patient), or other information specific to an
institution (e.g., standards, standard operating procedures,
policies, etc).
[0063] Another medical resource illustrated in FIG. 1 includes
medical decision support (CDS) service 180. CDS service 180 is a
resource for assessing medical conflicts. For example, CDS service
180 may include information about the suitability and compatibility
of medications, medical/surgical procedures, and/or diagnostic
tests, etc., and interactions therewith. For medications, CDS
service 180 may include information regarding dosages, forms and
routes of administration, compatibility between different types of
medication, compatibility of medications with particular allergies,
etc. With respect to procedures, CDS service 180 may include
information regarding compatibility between certain procedures and
pre-existing medical conditions, compatibility with medications,
etc. CDS service 180 may provide other types of information in this
regard, as the aspects are not limited in this respect. In some
systems, CDS functionality may be provided in whole or in part by
EHR system 170, or EHR system 170 may integrate CDS service 180 or
be capable of utilizing CDS service 180 as a resource, and the
illustrated configuration in FIG. 1 is merely exemplary. It should
be appreciated that CDS service 180, if implemented, may include
any amount of information including one, some or all of the types
of information described above.
[0064] Another medical resource illustrated in FIG. 1 includes
order fulfillment system 190. Order fulfillment system 190 may
include one or more systems capable of fulfilling orders submitted
in an applicable electronic format. Order fulfillment system 190 in
FIG. 1 denotes any one or combination of available order
fulfillment systems or resources capable of fulfilling applicable
orderable items. The one or more order fulfillment systems may be a
single system or multiple systems and is/are illustrated together
for convenience. According to some embodiments, order fulfillment
system 190 includes a Computerized Physician Order Entry (CPOE)
system capable of receiving and processing electronic orders for
one or any combination of medications, laboratory tests, radiology
screenings, surgical or other medical procedures and/or other
planned treatment action items.
[0065] As discussed above, some embodiments include a virtual
medical assistant configured to identify instruction to order one
or more orderable item specified in free-form instruction, and
assist the medical professional in placing the order. In this
respect, the virtual medical assistant may be configured to access
order fulfillment system 190 to facilitate the dispatch of orders
received by the virtual medical assistant from the medical
professional. To facilitate electronic ordering, the virtual
medical assistant may be configured to utilize any industry
standard electronic interfaces to connect with and communicate with
order fulfillment system 190, or may use any specialized or
proprietary interfaces or formats needed to utilize one or more
order fulfillment systems. For example, a virtual medical assistant
may generate an order to be presented as an electronic message in
accordance with the HL7 standard or in any other suitable manner.
Examples of industry standard electronic interfaces include
ePrescribing or CPOE for prescriptions, the CPOE system for
ordering procedures, and any of numerous systems for ordering lab
tests, order management scheduling systems, etc.
[0066] The exemplary system illustrated in FIG. 1 also includes
medical information retrieval resource 155 that may be utilized by
a virtual medical assistant in accordance with some embodiments.
Medical information retrieval resource may include one or more
resources that are available to provide or procure medical
information and may include one or more medical information
repositories and/or one or more search facilities configured to
respond to queries to locate medical information stored in the one
or more medical information repositories. According to some
embodiments, a virtual medical assistant may utilize medical
information retrieval resource 155 to facilitate responding to
medical professional inquiries, to identify one or more conflicts,
to infer information needed to perform one or more medical tasks,
or for one or more other purposes pursuant to assisting a medical
professional. However, a virtual medical assistant need not be
configured to access such a medical information retrieval resource
155, as the aspects are not limited in this respect.
[0067] As discussed above, the inventors have recognized benefits
of a virtual medical assistant configured to assist a medical
professional in performing one or more medical tasks that the
medical professional may need assistance in performing including,
but not limited to, any one or combination of accessing patient
information, preparing and/or submitting an order for an orderable
item (e.g., preparing and/or submitting a medication order, a
procedure order, a lab order, a diagnostic test order, an
appointment order, a referral order, a consultation order, etc.),
capturing, evaluating and/or archiving dictation associated with a
patient encounter, scheduling appointments, etc. The virtual
medical assistant may be configured to interact with the medical
professional in real-time to assist in the completion of medical
tasks desired by the medical professional. Such interactivity
allows for the virtual medical assistant to respond to instructions
received from the medical professional to obtain missing
information, disambiguate instruction, identify desired medical
tasks, identify conflicts and alert or warn the medical
professional regarding the conflicts, confirm an understanding of
instructions received, etc. Additionally, interaction with the
medical professional may allow the virtual medical assistant to
infer information needed to perform one or more medical tasks, for
example, using information in previously received instruction,
information obtained in performing one or more medical tasks in
response to previously received instruction, information in
electronic medical records and/or medical information accessed over
a network. Some exemplary techniques with respect to implementing a
virtual medical assistant according to some embodiments are
described in further detail below.
[0068] FIG. 2 is a flow chart illustrating a method performable by
a virtual medical assistant to assist a medical professional,
wherein the virtual medical assistant is implemented, at least in
part, by at least one processor of a host device, according to some
embodiments. The host device may be a mobile device, for example,
the medical professional's smart phone having a virtual medical
assistant application installed as an application or "app" on the
smart phone, or may be any other suitable device capable of hosting
a portion of a virtual medical assistant (e.g., a virtual medical
assistant application). The host device may include one or more
wireless transceivers capable of communicating with at least one
network and/or external device so that the virtual medical
assistant application can access one or more network resources for
which the virtual medical assistant application is configured to
cooperate with.
[0069] In act 210, free-form instruction is received by the virtual
medical assistant from the medical professional (e.g., via
interface functionality on a mobile device or other interface
device). For example, the medical professional may speak the
free-form instruction and the virtual medical assistant application
may receive the free-form speech instruction via one or more
microphones coupled to the mobile device. Alternatively, the free
form instruction may be received via another input component (e.g.,
via one or more other input mechanisms coupled to a mobile device).
For example, the medical professional may type the free form
instruction via a key pad or touch sensitive display screen on a
mobile device and the virtual medical assistant application may
receive the free-form instruction as a free-form textual
instruction. The free-form instruction may be received in any other
suitable way, as the aspects are not limited in this respect (e.g.,
free-form instruction may be received via an interface component
separate from the host device on which the virtual medical
assistant application is at least partially implemented).
[0070] In act 220, the received free-form instruction is provided
by the virtual medical assistant application for processing to
identify one or more medical tasks specified in the free-form
instruction by the medical professional. According to some
embodiments, after performing pre-processing of the free-form
instruction (if any such pre-processing is needed or desirable),
the virtual medical assistant application may transmit the
free-form instruction to be processed by one or more resources
connected to a network accessible by the host device. For example,
the virtual medical assistant application may provide the free-form
instruction to a medical understanding resource to analyze the
free-form instruction to determine its content and meaning, for
example, by utilizing an ASR component, NLU/CLU component and/or a
dialog component to process the free-form speech to identify a
medical task and/or one or more parameters associated with the
identified medical task. According to some embodiments, other
network resources are utilized to assist in identifying one or more
medical tasks specified in the received free-form instruction.
[0071] In some instances, processing the free-form instructions
results in identification of one or more impediments to performing
the identified at least one medical task. The term "impediment"
refers herein to any condition that generally impedes, complicates
and/or otherwise gives rise to questions or concerns with respect
to performing at least one identified medical task. Examples of
impediments are missing information needed (or optionally used) to
perform at least one medical task or an aspect of at least one
medical task, ambiguity with respect to performing at least one
medical task, conflicts associated with at least one medical task,
uncertainty as to the nature of one or more specified medical
tasks, etc. As should be appreciated from the examples given above,
an impediment need not operate to absolutely prevent or bar the at
least one medical task from being performed, but need only give
rise to a question or concern in this respect (e.g., a free-form
instruction may include sufficient information to complete the
order, but a compatibility or suitability conflict may have been
identified). However, some impediments may indeed prevent
performance of the identified medical task (e.g., missing
information required to complete an order for an orderable item may
be identified). It should be appreciated that in some instances
multiple impediments may be identified and in other instances no
impediments may be identified.
[0072] In act 230, at least one first response is provided to the
medical professional by the virtual medical assistant by, for
example, the virtual medical assistant application presenting the
at least one first response to the medical professional via the
host device (e.g., a mobile device) prior to the at least one
identified medical task being performed (or by presenting the at
least one first response via any interface device utilized by the
medical professional). For example, the at least one first response
may be displayed to the medical professional on a display of a
mobile device and/or the at least one first response may be
presented as audio to the medical professional using speech
synthesis techniques and rendering the speech via one or more
speakers coupled to the mobile device. According to some
embodiments, the at least one first response may be both visually
and audibly rendered to the medical professional, or a single mode
of presentation may be used, as the aspects are limited for use
with any particular presentation or rendering technique.
[0073] The one or more techniques for rendering information to the
medical professional may depend on the interface device utilized by
the medical professional and/or the environment of the medical
professional. For example, the display capabilities of a mobile
device may be limited in which case an audio presentation may be
preferable, or the medical professional may be utilizing the
virtual medical assistant in an environment where the focus of the
medical professional is trained elsewhere (e.g., while driving,
while performing an examination or procedure, etc.) such that an
audio presentation may be preferred (with or without a
corresponding visual rendering of the at least one first response).
According to some embodiments, the interface utilized by the
medical professional may include only voice capabilities such that
interaction between the medical professional and the virtual
medical assistant is conducted audibly. In some circumstances, the
medical professional may be using the virtual medical assistant in
circumstances where audio presentation may be inappropriate or
difficult to hear or understand, such that a visual rendering of
the at least one response alone may be preferable. However,
responses from the virtual medical assistant may be provided and
rendered in any suitable way, format or mode, as the aspects are
not limited in this respect.
[0074] The at least one first response, however presented, may be
any response deemed appropriate by the virtual medical assistant in
connection with the identified at least one medical task. For
example, the virtual medical assistant may identify one or more
impediments to performing the at least one medical task and
formulate (e.g., using a dialog component) the at least one first
response to include one or more inquiries provided to the medical
professional with respect to the one or more identified
impediments. In this respect, the at least one first response may
be formulated to include a request that the medical professional
provide at least some missing information needed to perform at
least one medical task or an aspect of at least one medical task, a
request for information to assist in resolving one or more
ambiguities associated with performing at least one medical task, a
question about what type of medical task the medical professional
would like performed, one or more alerts, warnings and/or cautions
associated with at least one identified medical task, etc.
[0075] The at least one first response alternatively may be
unrelated to any impediment. For example, the virtual medical
assistant may identify one or more medical tasks without detecting
any impediment to performing the at least one medical task, but
nevertheless, the medical virtual assistant may want to request
confirmation from the medical professional before initiating
performance of the identified medical task. Thus, the at least one
first response may be formulated to include a statement regarding
an understanding of the free-form instruction and/or a recitation
of the at least one medical task that the virtual medical assistant
intends to initiate performance of and/or any details that are
relevant (e.g., one or more parameters associated with the task).
The at least one first response may be formulated to include a
request that the medical professional confirm that the
understanding is correct and/or a request for confirmation to
proceed with performance of an identified medical task.
[0076] In act 240, the virtual medical assistant receives
information from the medical professional responsive to the at
least one first response. For example, the medical professional may
respond by providing missing information, disambiguation
information, guidance with respect to at least one medical task,
confirmation to proceed, verification as to the understanding of
the free-form instruction, cancellation of the free-form
instruction, modification to the free-form instruction, etc. The
information from the medical professional may be received by the
virtual medical assistant application via any suitable interface
device (e.g., using a mobile device) using any of the input
mechanisms described above in connection with receiving free-form
instruction.
[0077] In act 245, the virtual medical assistant may assess whether
the at least one medical task should be performed based, at least
in part, on the information received from the medical professional.
In particular, the virtual medical assistant may evaluate the
information to assess whether the at least one medical task should
be performed. For example, the virtual medical assistant system may
check to see if all of the information needed to perform a
requested medical task has been obtained, or whether further
information is needed, whether ambiguities are resolved or further
clarification is required, whether the medical task to be performed
has been correctly identified, whether any conflicts have been
cleared or the instruction modified to remove the conflict, whether
the medical professional has confirmed that a requested medical
task should be performed, etc. If one or more impediments have been
sufficiently overcome and/or the medical professional has confirmed
the understanding of the free-form instruction and authorized the
virtual medical assistant to proceed, the virtual medical assistant
may initiate performance of at least on medical task identified in
the free-form instruction (act 250).
[0078] If the virtual medical assistant deems that further
information is needed to perform an identified medical task, any
one or combination of acts 210-240 may be again performed. For
example, the virtual medical assistant may return and await further
free-form instruction from the medical professional (e.g., in the
event the medical professional decided to not proceed with an
identified medical task). Alternately, the virtual medical
assistant system may again process the free-form instruction in
view of the further information and/or one or more further
responses may be generated and provided to the medical professional
by the virtual medical assistant to invoke the medical professional
to provide further information regarding the free-form instruction.
The virtual medical assistant system may also assess whether
further information may now be inferred based on the additional
information received from the medical professional.
[0079] It should be appreciated that this type of dialog between
the medical professional and the virtual medical assistant may
continue until it is determined that one or more medical tasks are
ready to be performed and/or authorized by the medical
professional. In this manner, the virtual medical assistant may
interact with the medical professional in real-time to assist the
medical professional in performing medical tasks and/or in
performing medical tasks on behalf of the medical professional. It
should be further appreciated that identifying at least one medical
task and interacting with the medical professional may include
other acts not explicitly described in connection with the method
illustrated in FIG. 2. Furthermore, no one or any particular
combination of acts are required to be performed, nor do acts need
be performed in the order described in connection with FIG. 2, or
in any particular order, as the aspects are not limited in this
respect. Instead, method 200 merely provides an example of an
interaction between a medical professional and a virtual medical
assistant system according to some embodiments.
[0080] As one example of performing method 200, a medical
professional may provide the following free-form instruction in
connection with a patient planning an ascent of Mount Kilimanjaro
with a night stay in the crater: "Start John Doe on Diamox, 250 mg
pills taken once daily beginning two days prior to ascent." The
virtual medical assistant may process the free-form instruction and
recognize that Diamox is the brand name for the active ingredient
acetazolamide. The virtual medical assistant may access John Doe's
patient records and determine that the patient is allergic to
sulpha-based drugs. In response, the virtual medical assistant may
formulate a response including an alert to the medical professional
to this detected drug-allergy interaction by providing the
following response: "Patient John Doe is allergic to sulpha-based
drugs. Do you want to continue?" In response, the medical
professional may decide to change the prescription by providing the
following instruction: "Okay, order Dexamethasone instead."
[0081] The virtual medical assistant may receive the further
instruction from the medical professional, process the instruction
to ascertain that the medical professional has changed the
prescription to the steroid Dexamethasone. In processing the
further free-form instruction, the virtual medical assistant may
ascertain that the medical professional has failed to provide a
number of details needed to complete and dispatch the medication
order (e.g., dose, frequency, route, etc.). The virtual medical
assistant may provide a response to the medical professional as
follows: "Change medication order for patient John Doe to
Dexamethasone. To complete this order, please provide prescription
details." The medical professional may respond with the instruction
"4 mg pills taken orally twice a day starting the day before
ascent." This example discourse may continue until the virtual
medical assistant has obtained the information needed to complete
and dispatch the medication order, including, according to some
embodiments, a recitation of the completed order and a request for
the medical professional to confirm the order and/or for the
medical professional to provide any needed verification
information.
[0082] As one example of performing method 200 wherein the virtual
medical assistant requests further guidance regarding the medical
task(s) to be performed, a medical professional may provide the
following free-form instruction: "I need to consult with Dr. Jones
to discuss an ultrasound guided biopsy for kidney mass and
surrounding lymph nodes." The virtual medical assistant may receive
this free-form instruction, process the instruction and determine
that clarification on the intended directives of the medical
professional is needed. For example, the virtual medical assistant
may want clarification as to whether the medical professional wants
to schedule a consultation with Dr. Jones, order the biopsy
procedure, or both. In response, the virtual medical assistant may
provide a response to the medical professional as follows: "Would
you like me to schedule a consultation with Dr. Jones?" If the
medical professional responds in the positive, the virtual medical
assistant can initiate processes to set-up the consultation. The
virtual medical assistant, having identified multiple potential
tasks to be performed, may follow-up with the inquiry: "Would you
like me to order the ultrasound biopsy for kidney mass and
surrounding lymph nodes?" to which the medical professional may
reply: "No, let's wait until after the consultation." The virtual
medical assistant may note the date and time of the scheduled
consultation and set up an internal reminder to follow-up after the
consultation has taken place. After the consultation has occurred,
the reminder set by the virtual medical assistant may trigger the
virtual medical assistant to provide the following inquiry to the
medical professional "Would you like to order the ultrasound guided
biopsy of the kidney mass and surrounding lymph nodes pursuant to
your consultation with Dr. Jones?" In this way, the virtual medical
assistant can proactively assist the medical professional in
his/her practice.
[0083] As an example of performing method 200 wherein the at least
one response is not related to an impediment, a medical
professional may provide the following free-form instruction to the
medical professional: "Set up a full blood panel for Jane, with RBC
and WBC count and urine test." The virtual medical assistant may
receive and process the free-form instruction and ascertain that
the medical professional would like to order lab tests for Jane Doe
and may respond with the following: "Laboratory test for patient
Jane Doe: Full blood panel including red blood cell and white blood
cell count and urine test. Any special instructions?" The medical
professional may respond with the following: "Yes, 12-hour fast
required." The above exemplary interaction may facilitate the
provision of a comprehensive order for a lab test for the patient.
According to some embodiments, the virtual medical assistant may
follow-up the above dialog to provide an additional response that
recites the order that the assistant intends to place (including
the special instructions received) and request that the medical
professional confirm the order.
[0084] It should be appreciated that the above exemplary dialogs
between the medical professional and a virtual medical assistant
are merely exemplary of the types of discourse that a virtual
medical assistant may implement in assisting a medical
professional. Any suitable exchange between a medical professional
and a virtual medical assistant may be performed to assist a
medical professional in performing one or more medical tasks, as
the aspects are not limited to any particular type or form of
interaction. Some further examples are described below for
illustration.
[0085] Method 200 described above in connection with FIG. 2
illustrates certain discourse capability of a virtual medical
assistant system that allows the virtual medical assistant to
interact with the medical professional to probe, inquire, question,
request, confirm and/or verify in connection with free-form
instruction received from the medical professional. According to
some embodiments, a virtual medical assistant system may be
implemented that can infer information in connection with free-form
instruction provided by a medical professional. In this manner, the
virtual medical assistant may be able to handle and act upon
free-form instruction that may be incomplete, ambiguous,
inconsistent and/or that gives rise to one or more conflicts in
connection with one or more medical tasks specified in received
free-from instruction.
[0086] FIG. 3 includes a flow chart illustrating a method that may
be performed by a virtual medical assistant to assist a medical
professional by inferring at least some information in connection
with free-form instruction received from the medical professional.
Acts 310 and 320 may be similar in some respects to acts 210 and
220 described above in connection with method 200 illustrated in
FIG. 2 in that free-form instruction may be received from a medical
professional by the virtual medical assistant and processed to
identify at least one medical task to be performed. For example,
free-form instruction may be received from the medical professional
via a host device on which a virtual medical assistant application
is operating (or via other interface component(s)), and the
free-form instruction may be transmitted over a network to be
processed by one or more components forming, at least in part, the
virtual medical assistant to identify any medical tasks specified
in the free-form instruction.
[0087] In act 330, at least one impediment to performing a medical
task ascertained from the free-form instruction is identified. For
example, the virtual medical assistant may process the free-form
speech (e.g., using a medical understanding resource) to understand
the content of the free-form speech to ascertain what the medical
professional is instructing the virtual medical assistant to do or
assist with. In doing so, one or more impediments to carrying out
the directives of the medical professional may be identified. For
example, the virtual medical assistant may determine that
information needed to perform a requested medical task is missing,
may identify one or more ambiguities in the medical professional's
instruction, or may otherwise detect one or more conditions that
impede, complicate and/or raise uncertainty with respect to
performing one or more identified medical tasks.
[0088] In act 340, at least some information is inferred to assist
in overcoming the identified at least one impediment. For example,
the virtual medical assistant may infer information based on
information available to one or more components of the virtual
medical assistant, or available via one or more network resources
accessible by the virtual medical assistant. According to some
embodiments, the virtual medical assistant evaluates previously
received instruction from the medical professional and/or
information obtained responsive to previously received instruction
to infer the at least some information.
[0089] To illustrate an example of inferring at least some
information regarding one or more impediments to performing a
medical task, a medical professional may have provided the virtual
medical assistant the instruction: "Show me John Doe's chart." In
response to the instruction, the virtual medical assistant may have
initiated processes to obtain John Doe's patient history or other
relevant records from, for example, an EHR system accessible by the
virtual medical assistant, and presented at least a portion of John
Doe's "chart" to the medical professional. In response or
subsequent to receiving patient information for John Doe, the
medical professional may speak the free-form instruction "Please
refill prescription." The virtual medical assistant may process the
instruction and identify that the medical professional would like
to order a medication, but also ascertain that there is information
missing that is needed to perform the requested medical task. In
particular, the patient name is not explicit in the instruction,
and the medication, dose, frequency, administration, quantity, etc.
are missing entirely.
[0090] The virtual medical assistant may infer one or more of these
missing parameters from previous instruction or information
obtained from performing tasks responsive to previous instructions.
For example, the virtual medical assistant may recognize that the
medical professional is likely referring to John Doe due to the
previous instruction to "Show me John Doe's chart." In addition,
the virtual medical assistant may understand that the medical
professional is ordering a refill on a prescription and that the
prescription information may be present in the patient history
obtained in response to the instruction to obtain John Doe's chart,
or otherwise obtainable from an EHR system storing records for John
Doe. Accordingly, the virtual medical assistant may infer at least
some of this information from the electronic medical record for
John Doe.
[0091] In the example above, in attempting to infer at least some
information in connection with the prescription that the medical
professional intends to fill, the virtual medical assistant may
discover from John Doe's EMR that John Doe has multiple medication
prescriptions. The virtual medical system may provide a response to
the medical professional to obtain clarification (e.g., the virtual
medical assistant may perform functionality similar to or the same
as act 230 in method 200 illustrated in FIG. 2). For example, the
virtual medical assistant may present to the medical professional
the multiple prescriptions that were located (or an identifying
shorthand such as the medication name) along with a visual and/or
audible response such as "John Doe has multiple prescriptions.
Please select the prescription you would like to refill." The
medical professional may then identify the desired prescription to
allow the virtual medical assistant to proceed with the medication
order. Otherwise, if John Doe has only a single prescription, the
virtual medical assistant may infer that this is the prescription
the medical professional intends to refill and obtain the needed
information from John Doe's electronic medical record.
[0092] Prior to performing the medical task of ordering the
medication, the virtual medical system may present its
understanding of the instruction and the details of the
prescription refill (e.g., the parameters of the medication order)
to the medical professional to confirm that the instruction was
understood correctly and to confirm that any inferred information
is correct. In the above example, the virtual medical assistant may
present to the medical professional the confirmation "Medication
order for patient John Doe, Albuterol inhaler to be taken by oral
inhalation as needed. Please confirm medication order." If the
medical professional confirms (e.g., act 345 results in a yes), the
virtual medical assistant may initiate processes to submit the
medication order, such as generating a complete medication order
and submitting it to a CPOE system.
[0093] The virtual medical assistant may also infer one or more
parameters based on medical knowledge obtained from one or more
medical information resources available to the virtual medical
assistant (e.g., via one or more information retrieval resources).
For example, the medical professional may say "Start John Doe on
Compazine 2-3 times daily as needed for nausea." While the medical
professional does not explicitly state that the medication should
be taken orally, the virtual medical assistant may infer oral
administration by ascertaining that Compazine is the brand name for
the oral form of the drug prochlorperazine. In responding to the
medical professional (e.g., in a confirmatory response or, for
example, to inquire as to the dose), the virtual medical assistant
may include this inference in its recitation of the medication
order provided to the medical professional (e.g., "Medication order
for patient John Doe. Oral administration of Compazine in pill form
2-3 times a day as needed for nausea. What dose of the
prochlorperazine should be administered?").
[0094] As a further example, free-form instruction may mention an
"IV," (i.e., the medical route intravenous), and the virtual
medical assistant may infer that the medication should be
administered in a liquid form (which may be a parameter needed to
complete an order for a medication) so that in some embodiments
described herein, the liquid form can be included in the order for
the medication because it is implied. As another example, if the
free-form instruction indicates that medication should be
administered "stat," the virtual medical assistant may infer that
the medication needs to be administered immediately, which may
enable the virtual medical assistant to populate the start time
field in an order form for the medication.
[0095] The virtual medical assistant may also infer information
based on its knowledge of the medical professional and/or the
medical professional's practice. For example, the medical
professional may say "Jane Doe needs a CT scan of chest and neck,
with co." The virtual medical assistant may receive the free-form
instruction, process the free-form instruction to ascertain that
the medical professional would like to order a diagnostic test.
While the medical professional does not specify a radiology site to
perform the diagnostic test, the virtual medical assistant may
infer the radiology site based on the medical professional's
primary affiliation with an institution, previous indication of a
site to perform radiology tests, or other information available to
the virtual medical assistant that would suggest one site over
another. The virtual medical assistant may present its inference to
the medical professional in a confirmatory response or in a
response to obtain any further information needed to complete and
dispatch the order for the diagnostic test.
[0096] Method 300 described above in connection with FIG. 3
illustrates certain inference capabilities of a virtual medical
assistant system according to some embodiments that allow the
virtual medical assistant to perform medical tasks, conduct actions
on behalf of or otherwise assist a medical professional in
circumstances where medical professional instruction is incomplete,
ambiguous or otherwise gives rise to questions, by inferring at
least some information about the one or more medical tasks to be
performed. As examples above illustrate, techniques for inferring
information may be used alone or in combination with techniques for
interacting with the medical professional through discourse (e.g.
providing responses to the medical professional in connection with
received free-form instruction) to provide a system that may allow
a medical professional to conduct his/her practice more
efficiently, conveniently and/or safely, though achieving such
benefits is not a requirement of techniques described herein, or of
any implementation.
[0097] As discussed above, according to some embodiments, a virtual
medical assistant may be configured to perform one or more tasks
that may assist in a medical professional's practice of medicine.
Exemplary tasks include, but are not limited to, sending messages
(text, email, voice, etc.), scheduling meetings, capturing and/or
processing medical dictation, accessing medical information in EMRs
and/or locating and presenting information from and EHR system,
preparing and submitting orders (e.g., medication orders, lab
orders, procedure orders, diagnostic test orders, consultation
orders, referral orders, etc.), providing medical decision support
through medical search and alerts, etc. It should be appreciated
that a virtual medical assistant may implement one or any
combination of the above exemplary tasks, or others not explicitly
described herein, and there are no requirement that a virtual
medical assistant perform any particular task.
[0098] According to some embodiments, a virtual medical assistant
can operate as an interface to an EHR system to access EMRs to
obtain patient information requested by the medical professional
and/or to assist in the performance of medical tasks requested by
the medical professional. The virtual medical assistant may
maintain patient context across multiple instructions from the
medical professional to facilitate a meaningful presentation to the
medical professional and to allow the virtual medical assistant to
infer information in connection with medical professional
instruction.
[0099] In this respect, the virtual medical assistant may operate
as an EMR portal to retrieve patient information for review by the
medical professional and/or to assist in performing one or more
medical tasks including, but not limited to, any one or combination
of obtaining and presenting a list of current patients, selecting a
patient for more detailed viewing, searching for patients that
match specified criteria and obtaining information on selected
patients. For selected patients, more detailed information may be
obtained via the virtual medical assistant. For example, the
virtual medical assistant may present a patient chart (or portion
thereof) or other patient data from the associated EMRs such as
current problem or disorder list, current medication list, allergy
list, current order, patient history notes such as medical notes,
consult notes, operative reports, surgical, pathology, radiology
reports, progress notes, treatment summaries, discharge summaries,
diagnostic test results, imaging results, etc. In this respect, the
virtual medical assistant may be configured to navigate a given EHR
system and individual EMRs to obtain patient information requested
by the medical professional.
[0100] As discussed above, some embodiments include a virtual
medical assistant configured to assist a medical professional in
electronically ordering an orderable item. For example, the virtual
medical assistant may receive free-form instruction from the
medical professional that includes instruction to place an order,
and may process the free-form instruction to prepare, complete
and/or dispatch the order, which may include making one or more
inferences about the order or orderable item and/or interacting
with the medical professional to obtain further information in this
regard. The virtual medical assistant may be configured to
interface with an order fulfillment system to dispatch the order
once the order is complete.
[0101] According to some embodiments, a virtual medical assistant
may be configured to understand multiple types of orders for any
one or combination of orderable items including in-patient orders
for medication in a medical facility (e.g., a hospital) and
out-patient medication orders (e.g., prescriptions), orders for
medical procedures (e.g., an MRI, a CT scan or any of numerous
other types of procedures), orders for laboratory work or other
tests, referral orders to other medical professionals (e.g., other
physicians or specialists, physical therapy, etc.), communication
orders (e.g., orders for instructions, counseling, guidance or
training for any of numerous topics such as smoking cessation or
dietary counseling), etc.
[0102] As discussed above, in accordance with some embodiments
described herein, a virtual medical assistant may process free-form
instruction received from a medical professional using a medical
understanding resource that includes an NLU/CLU component to
identify the order specified in the free-form instruction. The
NLU/CLU component may identify one or more parameters in the order,
or may identify information that may be used as a basis for
determining or obtaining parameters in the order. The medical
understanding resource (e.g., a dialog component used by the
medical understanding resource) may have an understanding of the
parameters that need to be specified to generate an order of the
applicable type. For example, the medical understanding component
(or other portion of the virtual medical assistant) may maintain a
representation of each order type with which the virtual medical
assistant is configured to assist that includes a field for each
parameter required and/or optional in a connection with placing the
corresponding order.
[0103] If the free-form instruction does not fully specify all of
the parameters for an order of the orderable item, in accordance
with some embodiments, missing parameters can be inferred by the
virtual medical assistant using information from previous
instructions, or information from resources available to the
virtual medical assistant, and/or obtained through dialog with the
medical professional. Exemplary parameters for ordering a
medication may include the name of the medication (e.g., a
commercial medication or drug name or a generic name), a dose, a
route, a frequency, a duration and an indication (e.g., specifying
a symptom, condition, disease, etc. that indicates that the
medication should be taken). Exemplary parameters for a procedure
may include an identification of the type of the procedure, as well
as a location and date for the procedure. Exemplary parameters for
a laboratory procedure or test may include an identification of the
type of procedure or test, a date, a frequency if it is to be
repeated, and any special instructions regarding the laboratory
procedure or test. Exemplary parameters for ordering instructions,
counseling, guidance, training, etc. may vary depending upon the
nature of the communication order and whether it is to be given
verbally or via some reference materials (e.g., written leaflets or
other material to be provided to the patient). Parameters for a
follow-up visit may include information identifying the type of
visit (e.g., is it a special purpose visit to address something in
particular, a general check-up, etc.), a date and any special
instructions.
[0104] When an order is complete, a virtual medical assistant may
generate an order and dispatch the order for electronic
fulfillment. As discussed above, a virtual medical assistant may
interface with a CPOE or other order fulfillment system to
automatically place orders for prescriptions, laboratory tests,
radiology screenings, surgical or other medical procedures and/or
other planned treatment action items, based on processing free-form
instruction to obtain parameters needed to complete the order. In
accordance with some embodiments, the virtual medical assistant may
employ one or more coding systems to facilitate placing an order in
its proper form. Such systems can be used to normalize terminology
and/or provide one or more codes for an orderable item that may be
required by a particular order fulfillment system. For example, a
medication with a name "X" may be orderable in a particular order
fulfillment system only by using a different name or code "Z". A
coding system may use one or more coding standards (e.g., RxNorm
for medications, LOINC for laboratory and diagnostic orders, etc.),
and the virtual medical assistant may be configured to utilize such
systems to place an order in the correct form prior to dispatching
the order to the appropriate order fulfillment system.
[0105] As discussed above, in accordance with some embodiments, a
virtual medical assistant may perform analysis on free-form
instruction received from a medical professional to detect and
provide an alert for potential conflicts in connection with the
order. This can be performed in any suitable manner, some examples
of which were discussed above. For example, when the orderable item
is a medication, the dosage may be checked by the virtual medical
assistant to determine if it is between a minimum and maximum
range, and/or its form and route may be checked for consistency. As
an another example, recent laboratory results for the patient may
be checked to ensure that the medication is compatible with them
(e.g., the patient's cholesterol level). As another example, drug
interactions may be checked to ensure that the medication being
ordered is compatible with medications on the patient's current
medication list (drug-drug interaction). As yet another example, a
check may be made for a medication to ensure that it is compatible
with any allergies in the patient's medical history (drug-allergy
interaction). As another example, drug interactions may be checked
to assess whether a particular drug is safe for use in connection
with a medical condition (drug-condition interaction).
[0106] When the orderable item is a procedure, the virtual medical
assistant may check to see if the procedure is compatible with any
medication taken by the patient (procedure-drug interaction),
whether the procedure is safe for a patient with a given condition
(procedure-condition interaction), or may detect any other
conflicts or issues in connection with a specified procedure and a
given patient. It should be appreciated that conflicts with respect
to compatibility and/or suitability may be determined by a virtual
medical assistant in connection with any one or combination of
order types understandable by the virtual medical assistant, as the
aspects are not limited in this respect. The virtual medical
assistant can perform such determinations in any suitable manner,
including using one or more CDS systems that include information
about the suitability and compatibility of dosages, forms and
routes of medication, compatibility between different types of
medication, compatibility of medications with particular allergies,
etc.
[0107] Described below are a number of example interactions between
a medical professional and a virtual medical assistant via a mobile
device to illustrate various concepts and techniques related to
virtual medical assistance, according to some embodiments. The
examples are described in connection with the wireframes in FIGS.
4A-4H which schematically illustrate a virtual medical assistant
application installed on a mobile device 401. FIGS. 4A-4H show an
exemplary GUI 410 presented to a medical professional by the
virtual medical assistant application on a display of mobile device
401 to allow the medical professional to utilize virtual medical
assistant functionality, some examples of which are described below
in connection with FIGS. 4A-4H.
[0108] FIG. 4A illustrates GUI 410 presented by a virtual medical
assistant on a mobile device, in accordance with some embodiments.
The general GUI functionality allows for a tap and speak discourse
rendered in a chat-like presentation on mobile device 401. In
particular, dialog from the virtual medical assistant and from the
medical professional are displayed in dialog icons on a display of
mobile device 401. A microphone icon 413 is also presented on the
display that the medical professional can tap when the medical
professional would like to speak an instruction, which the medical
professional can speak in free-form. Spoken free-form instruction
is captured by the mobile device and received by the virtual
medical assistant application resident on the mobile device. The
received free-form instruction may then be processed by the virtual
medical assistant by, for example, providing the received free-form
instruction to one or more network resources via wireless
communication component 406.
[0109] FIG. 4A illustrates an exemplary introductory presentation,
for example, that may be presented when the medical professional
activates the virtual medical assistant (e.g., by launching the
virtual medical assistant application on the mobile device). In
this example, the virtual medical assistant application displays
dialog (and may also audibly render the dialog) indicating that
virtual medical assistant is ready to assist and provides guidance
to the medical professional by inviting the medical professional to
tap the microphone icon and tell the virtual medical assistant what
the medical professional would like done. Initially, a medical
professional may be unfamiliar with functionality provided by the
virtual medical assistant and the virtual medical assistant
application may be present a menu of exemplary tasks that the
virtual medical assistant can assist the medical professional with
or perform on behalf of the medical professional. To this end, the
medical professional may say "What can you do for me?" and in
response the virtual medical assistant application may present a
number of example tasks the virtual medical assistant can perform
by providing representative free-form instruction suitable to
instruct the virtual medical assistant to perform the associated
task, or begin the process of performing such a task. It should be
appreciated that the medical professional is not limited to
speaking the same words presented in the examples, as the virtual
medical assistant may be configured to handle free-form
instruction. The menu of examples is intended to give the medical
professional a flavor for how to interact with the virtual medical
assistant.
[0110] The exemplary tasks presented to the medical professional
are illustrated in FIG. 4B and include making a telephone call to
someone in the medical professional's telephone directory, sending
a message (e.g., a text or email) to a designated recipient,
requesting that the virtual medical assistant show the medical
professional's schedule, patient list or chart for a particular
patient, ordering a medication, taking dictation (e.g., to
memorialize the results of a patient encounter), etc. A given
virtual medical assistant may choose to present any set of possible
tasks that the virtual medical assistant can help with, or such
"help" functionality may not be implemented at all, as this feature
is used to illustrate possible functionality of a virtual medical
assistant.
[0111] FIG. 4C illustrates the results of the medical professional
requesting that the virtual medical assistant "show my patient
list." In particular, in response to the medical professional's
free-form instruction, the virtual medical assistant presents a
list of the medical professional's patients, which may be obtained
from a EHR system or other location that maintains this
information. As discussed above, a virtual medical assistant may
receive instructions from the medical professional in other ways
besides speech. FIG. 4D illustrates the result of the virtual
medical assistant responding to the medical professional selecting
a patient from the patient list, for example, via touch
capabilities of the display. In particular, the virtual medical
assistant application may interpret the medical professional
touching the display on the entry for a particular patient as
instruction to retrieve the chart for this patient, which the
virtual medical assistant may obtain by accessing an EHR system. It
should be appreciated that the medical professional could have
provided this instruction using other input mechanism such as, for
example, by providing free-form instruction by saying "show me
[patient name's] chart," or by typing such an instruction via
keypad.
[0112] Medical professionals often have busy schedules and may need
assistance in scheduling appointments and/or may need reminders to
perform certain tasks. An on-the-go medical professional may
remember a task that needs to be completed, such as following up
with a patient or medical staff in connection with a patient, but
may not at a given moment have the time to follow-up, or it may not
make logistic sense to do so at a given time. The medical
professional may utilize the virtual medical assistant to help
manage their busy schedules by assisting in scheduling appointments
or patient visits, setting up alerts and/or reminders, managing
conflicts on the calendar, etc.
[0113] FIG. 4E illustrates a virtual medical assistant helping a
medical professional set-up a reminder. In particular, in the
example shown in FIG. 4E, the medical professional speaks the
free-form instruction "on April 12.sup.th remind me to follow-up
with John Cartales," and the free-form instruction is received by
the virtual medical assistant for processing. The medical
professional does not specify the year, but the virtual medical
assistant may reasonably infer the year to be the current year, and
proceed to set-up a reminder for Apr. 12, 2013. The virtual medical
assistant may present the reminder to the medical professional and
request that the medical professional confirm, either by presenting
a touchable icon or other manually activated confirmation,
processing a further free-form speech instruction, or both.
Alternatively, the virtual medical assistant may set up the
reminder and report details of the completed task without asking
for confirmation from the medical professional.
[0114] FIG. 4F illustrates a medical professional utilizing
dictation functionality to capture free-form narration in
connection with a patient encounter. For example, the medical
professional may say "Please take dictation for Jane Doe," and the
virtual medical assistant may interpret this instruction as the
medical professional wishing to provide a narrative regarding a
patient encounter, such as a patient visit, with Jane Doe. The
virtual medical assistant may instruct the medical professional to
tap the microphone and beginning speaking the narrative. The
free-from narrative may be captured by the device, received by the
virtual medical assistant and processed to recognize the content of
the dictation. The recognized dictation may be provided back to the
user on the display via the virtual medical assistant application.
According to some embodiments, the virtual medical assistant may
merely receive the audio and transfer the audio to another system,
e.g., a system that archives audio dictation for transcription
either by a transcriptionist or by an offline batch automatic
recognition system. According to some embodiments, the recognized
dictation (e.g., recognized by ASR component 130) is transmitted to
an EHR system to be included in the patient's electronic medical
record. According to some embodiments, the recognized dictation is
provided to an NLU component to extract medical facts to add the
facts to the EHR system according to the structure of the database.
According to some embodiments, the virtual medical assistant may
infer some information about the dictation and provide this
information to populate fields in an EHR system (e.g., the virtual
medical assistant may infer the date from the statement "Regarding
Tuesday's visit . . . " and may also populate patient name
information by virtue of receiving the instruction "Please take
dictation for Jane Doe.").
[0115] FIGS. 4G and 4H illustrate an example dialog between a
virtual medical assistant, wherein the medical professional
utilizes the virtual medical assistant to order medication for a
patient. As shown in FIG. 4G, the medical professional speaks the
free-form instruction: "Order a medication for Joe Petro," which is
received by the virtual medical assistant application via the audio
input capabilities of the mobile device. The virtual medical
assistant application on the mobile device may then provide the
free-form instruction to a medical understanding resource of the
system for processing to recognize the medical professional's
speech and understand the speech content. In this example, the
virtual medical assistant understands that the medical professional
would like to order a medication and recognizes that additional
information is needed to complete the medical task of submitting a
medication order.
[0116] To begin the process of obtaining further information, the
medical virtual assistant formulates the response: "Here's your
medication order for Joe Petro. What medication?" and presents the
response to the medical professional. In this example, the virtual
medical assistant, in its response to the free-form instruction,
both queries the medical professional for further information and
presents to the medical professional a medication order filled out
with the information gleaned by the virtual medical assistant from
the free-form instruction, which in the first instance consists
primarily of the patient name and date of birth (which the virtual
medical system may have obtained by accessing a EHR system, or
other medical records system available to the virtual medical
assistant). Presenting the medication order to the medical
professional allows the medical professional to quickly confirm
that the virtual medical assistant understood the instruction and
has identified the correct patient so that the medical professional
feels comfortable proceeding.
[0117] In the example illustrated in FIGS. 4G and 4H, in response
to the query regarding the medication, the medical professional
responds by saying: "Medication name Laxix, dose is 40 mg, route is
IV, start at 3 pm today. Please confirm." The virtual medical
assistant receives the free-form speech instruction and processes
the speech to extract information needed to complete the medication
order. In this example, the virtual medical assistant extracts the
medication name, dose, route and time of administration and
presents this information on the medication order for the medical
professional to review. It should be appreciated that any of the
responses provided by the virtual medical assistant may be rendered
as audio either alone or to supplement a visual display of the
response. The virtual medical assistant application also presents a
confirmation icon to the medical professional so that the medical
professional can confirm that the medication order is correct.
While the medical professional explicitly requested confirmation,
the virtual medical assistant may automatically await confirmation
for any medical task where confirmation is appropriate and/or
required, thus facilitating performance of medical tasks with fewer
mistakes and avoiding medical error. The virtual medical assistant
may be configured to always request confirmation or may dispense
with confirmation for certain task unless the medical professional
requests confirmation, as the aspects are not limited in this
respect.
[0118] If the medical professional confirms that the medication
order is correct (e.g., by tapping the "Confirm" icon, the virtual
medical assistant may generate a medication order formatted to
conform to a target ordering system (e.g., a desired CPOE system)
and submit the order for processing by the CPOE system to complete
the medication order. In the example in FIGS. 4G and 4H, the
medical professional provided the needed information in a single
response. However, if the virtual medical assistant required
further information from the medical professional, the discourse
with the medical professional may continue until the information
needed to complete the order is received. Additionally, any other
impediment identified by the virtual medical assistant, such as
issues with drug interaction, dose or route incompatibility or any
questions the virtual medical assistant has in connection with the
medication order, may be worked out with the medical professional
via an appropriate dialog exchange between the medical professional
and the virtual medical assistant.
[0119] As discussed above, a virtual medical assistant may be
configured to dialog with a medical professional by rendering
responses (e.g., prompts, requests, inquiries, alerts, warnings,
etc.) either visually or audibly. With respect to audibly rendering
responses, the virtual medical assistant may employ speech
synthesis techniques, either over a network or using local
resources, to generate audio responses that can be rendered, for
example, using one or more speakers coupled to the device hosting
the virtual medical assistant or using a separate interface
component. Providing audio responses may be used alone, or in
combination with visual responses, or one or the other may be used
depending on the circumstances, environment or medical professional
preference.
[0120] Table 1 below illustrates an example discourse between a
medical professional and a virtual medical assistant essentially
conducted in audio to allow, for example, the exchange to take
place without requiring the medical professional to visually focus
on the device and/or in situations wherein the device being used by
the medical professional to interact with the virtual medical
assistant does not include display functionality, or the virtual
medical assistant does not have access to display functionality of
the device. In particular, the free-form instruction may be
provided by the medical professional using speech and the responses
from the virtual medical assistant may be rendered via audio using
speech synthesis techniques and/or prerecorded speech. Such an
audio mode may be preferable in certain circumstances such as when
the medical professional's visual focus is trained elsewhere (e.g.,
while driving, performing a procedure or otherwise attending to a
patient, etc.), when the display of the device hosting the virtual
medical assistant makes visual responses difficult or otherwise
problematic, or visual responses are not possible (e.g., the
interface used by the medical professional does not include display
capabilities such as a telephone call-in environment). An example
discourse for ordering medication may proceed as follows.
TABLE-US-00001 TABLE 1 EXAMPLE AUDIBLE DIALOG Medical professional:
Write a medication order for John Doe, give Furosemide 40 mg po
stat Provider: CPOE medication order info: Patient is John Doe,
Medication Name Furosemide, Dose 40 mg, Route is po, Priority is
stat, Start Time is Now. What Medication form? Medical
professional: Form is pill Provider: CPOE order information is
complete: Patient is John Doe, Medication Name Furosemide, Dose 40
mg, Route is po, administration is pill form, Priority is stat,
Start Time is Now. Medical professional: Correct, please proceed to
order. Provider: There is a potential allergic reaction alert with
this order. Do you want to change the order? Medical professional:
What is the problem? Provider: The patient is allergic to
penicillin, of category PCN Cephalospor for the reasons of
anaphylaxis. Medical professional: Override the alert, the patient
is not truly allergic. Provider: Okay, overriding the allergy alert
with overriding reason that the patient is not truly allergic.
Please speak your voice signature to override. Medical
professional: <Voice Signature> Provider: Thank you, CPOE
order for John Doe has been submitted, order number is #######. Do
you wish to make another order? Medical professional: Yes, order
Levofloxacin 5000 mg IV for 3 days everyday dose now, then routine
schedule times. Provider: CPOE medication order for John Doe,
Medication name Levoflaxin, dose 5000 mg, route IV, frequency every
day for 3 days. Start dose now, then routine schedule times. Please
confirm that this is correct and provide any special instructions?
Medical professional: Yes, correct. No special instructions.
Provider: The dose of 5000 mg is greater than recommended dose of
500 mg according to National Drug Data File recommendation. How
would you like to proceed? Medical professional: Override. The
reason is a ABN kidney function. Provider: Okay, overriding the
min/max dose alert with overriding reason of ABN kidney function.
Please say your voice signature to override. Medical professional:
<Voice Signature> Provider: Thank you. Medication order for
John Doe has been submitted, order number is #######. Do you wish
to make another order? Medical professional: Not right now.
[0121] It should be appreciated that the above discourse may also
be implemented using visual responses (e.g., using any of the
visual display techniques illustrated in FIGS. 4A-4H, or any other
suitable display technique or format) and/or medical professional
input may be received in ways other than speech (e.g., via keypad,
touch icons, etc.), as the example in Table 1 is merely
illustrative of one possible way of conducting a dialog with a
medical professional that may be suitable (or advantageous) in
certain situations, while visual display may be used and/or
preferred in other situations.
[0122] The example discourse illustrated in Table 1 demonstrates
the use of dialog and inference techniques to discourse with the
medical professional to understand and help complete the medical
task of placing a medication order. In particular, in multiple
instances, the virtual medical assistant provided responses to the
medical professional to obtain further information regarding the
medication order (e.g., request missing information such as
medication form, inquire as to special instructions, etc.) and/or
alert the medical professional to conflicts in connection with the
medication order (e.g., drug-allergy alert, min/max dose
alert).
[0123] In addition, in the example discourse illustrated in Table
1, the virtual medical assistant obtains information by inference
from previously received instruction and/or information obtained as
a result of previously received instruction. For example, the
medical professional's second medication order (e.g., Levofloxacin)
did not include the name of the patient but the virtual medical
assistant inferred that the patient was the same as for the
previous order.
[0124] A virtual medical assistant may be configured with security
measures such as any one or combination of password protection,
voice verification/authentication, signature
verification/authentication, etc. to verify and/or confirm that a
user providing free-form instruction is authorized to perform one
or more medical tasks specified in free-form instruction received
from the user. Any suitable security techniques may be utilized, as
the aspects are not limited in this respect.
[0125] An illustrative implementation of a computer system 500 that
may be used to implement one or more virtual medical assistant
techniques described herein is shown in FIG. 5. Computer system 500
may include one or more processors 510 and one or more
non-transitory computer-readable storage media (e.g., memory 520
and one or more non-volatile storage media 530). The processor 510
may control writing data to and reading data from the memory 520
and the non-volatile storage device 530 in any suitable manner, as
the aspects of the invention described herein are not limited in
this respect. Processor 510, for example, may be a processor on a
mobile device.
[0126] To perform functionality and/or techniques described herein,
the processor 510 may execute one or more instructions stored in
one or more computer-readable storage media (e.g., the memory 520,
storage media, etc.), which may serve as non-transitory
computer-readable storage media storing instructions for execution
by processor 510. Computer system 500 may also include any other
processor, controller or control unit needed to route data, perform
computations, perform I/O functionality, etc. For example, computer
system 500 may include any number and type of input functionality
to receive data and/or may include any number and type of output
functionality to provide data, and may include control apparatus to
perform I/O functionality.
[0127] In connection with identifying task(s) specified in
free-form instruction, one or more programs configured to receive
free-form instruction, process the instruction or otherwise execute
functionality described herein may be stored on one or more
computer-readable storage media of computer system 500. In
particular, some portions or all of a virtual medical assistant may
be implemented as instructions stored on one or more
computer-readable storage media. Processor 510 may execute any one
or combination of such programs that are available to the processor
by being stored locally on computer system 500 or accessible over a
network. Any other software, programs or instructions described
herein may also be stored and executed by computer system 500.
Computer system 500 may represent the computer system on a host
device on which a virtual medical assistant application is
installed and/or may represent the computer system on which any one
or combination of network resources are implemented (e.g., any one
or combination of components forming a virtual medical assistant,
or other network resource). Computer system 500 may be implemented
as a standalone computer, server, part of a distributed computing
system, and may be connected to a network and capable of accessing
resources over the network and/or communicate with one or more
other computers connected to the network (e.g., computer system 500
may be part of the network resources available to a virtual medical
assistant).
[0128] The terms "program" or "software" are used herein in a
generic sense to refer to any type of computer code or set of
processor-executable instructions that can be employed to program a
computer or other processor to implement various aspects of
embodiments as discussed above. Additionally, it should be
appreciated that according to one aspect, one or more computer
programs that when executed perform methods of the disclosure
provided herein need not reside on a single computer or processor,
but may be distributed in a modular fashion among different
computers or processors to implement various aspects of the
disclosure provided herein.
[0129] Processor-executable instructions may be in many forms, such
as program modules, executed by one or more computers or other
devices. Generally, program modules include routines, programs,
objects, components, data structures, etc. that perform particular
tasks or implement particular abstract data types. Typically, the
functionality of the program modules may be combined or distributed
as desired in various embodiments.
[0130] Also, data structures may be stored in one or more
non-transitory computer-readable storage media in any suitable
form. For simplicity of illustration, data structures may be shown
to have fields that are related through location in the data
structure. Such relationships may likewise be achieved by assigning
storage for the fields with locations in a non-transitory
computer-readable medium that convey relationship between the
fields. However, any suitable mechanism may be used to establish
relationships among information in fields of a data structure,
including through the use of pointers, tags or other mechanisms
that establish relationships among data elements.
[0131] Also, various inventive concepts may be embodied as one or
more processes, of which multiple examples have been provided. The
acts performed as part of each process may be ordered in any
suitable way. Accordingly, embodiments may be constructed in which
acts are performed in an order different than illustrated, which
may include performing some acts concurrently, even though shown as
sequential acts in illustrative embodiments.
[0132] All definitions, as defined and used herein, should be
understood to control over dictionary definitions, and/or ordinary
meanings of the defined terms.
[0133] As used herein in the specification and in the claims, the
phrase "at least one," in reference to a list of one or more
elements, should be understood to mean at least one element
selected from any one or more of the elements in the list of
elements, but not necessarily including at least one of each and
every element specifically listed within the list of elements and
not excluding any combinations of elements in the list of elements.
This definition also allows that elements may optionally be present
other than the elements specifically identified within the list of
elements to which the phrase "at least one" refers, whether related
or unrelated to those elements specifically identified. Thus, as a
non-limiting example, "at least one of A and B" (or, equivalently,
"at least one of A or B," or, equivalently "at least one of A
and/or B") can refer, in one embodiment, to at least one,
optionally including more than one, A, with no B present (and
optionally including elements other than B); in another embodiment,
to at least one, optionally including more than one, B, with no A
present (and optionally including elements other than A); in yet
another embodiment, to at least one, optionally including more than
one, A, and at least one, optionally including more than one, B
(and optionally including other elements); etc.
[0134] The phrase "and/or," as used herein in the specification and
in the claims, should be understood to mean "either or both" of the
elements so conjoined, i.e., elements that are conjunctively
present in some cases and disjunctively present in other cases.
Multiple elements listed with "and/or" should be construed in the
same fashion, i.e., "one or more" of the elements so conjoined.
Other elements may optionally be present other than the elements
specifically identified by the "and/or" clause, whether related or
unrelated to those elements specifically identified. Thus, as a
non-limiting example, a reference to "A and/or B", when used in
conjunction with open-ended language such as "comprising" can
refer, in one embodiment, to A only (optionally including elements
other than B); in another embodiment, to B only (optionally
including elements other than A); in yet another embodiment, to
both A and B (optionally including other elements); etc.
[0135] Use of ordinal terms such as "first," "second," "third,"
etc., in the claims to modify a claim element does not by itself
connote any priority, precedence, or order of one claim element
over another or the temporal order in which acts of a method are
performed. Such terms are used merely as labels to distinguish one
claim element having a certain name from another element having a
same name (but for use of the ordinal term).
[0136] The phraseology and terminology used herein is for the
purpose of description and should not be regarded as limiting. The
use of "including," "comprising," "having," "containing",
"involving", and variations thereof, is meant to encompass the
items listed thereafter and additional items.
[0137] Having described several embodiments of the techniques
described herein in detail, various modifications, and improvements
will readily occur to those skilled in the art. Such modifications
and improvements are intended to be within the spirit and scope of
the disclosure. Accordingly, the foregoing description is by way of
example only, and is not intended as limiting. The techniques are
limited only as defined by the following claims and the equivalents
thereto.
* * * * *